Microbial Fermentation of Glycerol For Production of Value-Added Compounds Under High Pressure

Universidade de Aveiro Departamento de Química
2018
MARIA JOÃO PINHO FERMENTAÇÃO MICROBIANA DE GLICEROL PARA

MOTA PRODUÇÃO DE COMPOSTOS DE VALOR-
ACRESCENTADO SOB ALTA PRESSÃO
MICROBIAL FERMENTATION OF GLYCEROL FOR

PRODUCTION OF VALUE-ADDED COMPOUNDS
UNDER HIGH PRESSURE
Universidade de Aveiro Departamento de Química
2018
MARIA JOÃO PINHO FERMENTAÇÃO MICROBIANA DE GLICEROL PARA

MOTA PRODUÇÃO DE COMPOSTOS DE VALOR-
ACRESCENTADO SOB ALTA PRESSÃO
MICROBIAL FERMENTATION OF GLYCEROL FOR

PRODUCTION OF VALUE-ADDED COMPOUNDS
UNDER HIGH PRESSURE
Tese apresentada à Universidade de Aveiro para cumprimento dos requisitos
necessários à obtenção do grau de Doutor em Química Sustentável, realizada
sob a orientação científica do Doutor Jorge Manuel Alexandre Saraiva,
Investigador Auxiliar do Departamento de Química da Universidade de Aveiro,
e da Professora Ivonne Delgadillo Giraldo, Professora Associada com
Agregação do Departamento de Química da Universidade de Aveiro.
Apoio financeiro do POCTI no âmbito Apoio financeiro da FCT e do FSE no

do III Quadro Comunitário de Apoio. âmbito do III Quadro Comunitário de
Apoio.
Dedico este trabalho aos meus pais, pelo incansável apoio ao longo de todas
as etapas da minha vida.
agradecimentos O Doutoramento foi um enorme desafio e um processo de aprendizagem e
desenvolvimento pessoal. Não teria sido possível sem a ajuda e o apoio de
inúmeras pessoas. Quero expressar a minha gratidão a todas elas, em
especial àquelas mencionadas em baixo.
Ao meu orientador, Professor Jorge Saraiva, pela disponibilidade, apoio e por

todos os conhecimentos partilhados. E, acima de tudo, pela positividade que
sempre transmitiu, até nos momentos mais difíceis, e que me ajudou a manter
a calma e a paciência. Obrigada por ter contribuído para meu desenvolvimento
profissional e também pessoal.
À minha co-orientadora, Professora Ivonne Delgadillo, pela orientação e

suporte científico, bem como pela transmissão de conhecimentos relevantes
para o desenvolvimento do trabalho.
À Professora Ana Gomes e ao Sérgio Sousa, pelo apoio no HPLC; ao

Professor Mário Simões, pela contribuição e ajuda no GC-MS; e ao Professor
Brian Goodfellow, pelo apoio no RMN e na análise multivariada dos resultados.
A todos os colegas do Innovate Group, pela colaboração no desenvolvimento

deste trabalho. Em particular, à Sofia, por ser a pessoa mais querida e
bondosa que conheci; à Patrícia, à Catarina e ao Ricardo Ferreira, por todas
as gargalhadas e momentos bem-dispostos. Todos vocês tornaram os meus
dias mais animados e felizes. Tenho que deixar um agradecimento especial à
Rita, a minha grande amiga e companheira de todas as lutas. Por tanta coisa
que nem consigo expressar em palavras. Existem poucas amizades assim,
espero que consigamos sempre preservá-la.
A toda a minha família e amigos, que nunca me deixam ter um momento mau
na vida.
Em particular, à Ana Alves e à Betânia, pela amizade sincera, independente do

tempo e da distância. À Priscila, ao Diogo, à Sara, ao Vítor e ao André,
companheiros de grandes aventuras “dolmáticas”. E a todos os amigos de
Oliveira, que são tantos que nem os posso enumerar aqui. Obrigada pela
amizade e por todos os momentos felizes e bem-dispostos, que foram
fundamentais para me manter motivada e conseguir concluir esta etapa.
À D. Gina, pelo carinho imenso, e por me tratar sempre como uma filha. À
Sarita, por me mostrar que as mulheres não se medem aos palmos. E pelo
fornecimento de gotinhas... Sem gotinhas não havia tese, ninguém consegue
estar tantas horas a escrever com os olhos secos.
agradecimentos E, claro, ao “núcleo duro” da minha família. Ao meu padrinho, à tia Lúcia, ao
João e à Mariana, por estarem sempre presentes e sempre disponíveis para
ajudar. Um agradecimento muito especial à avó Lena e ao avô Carlos, os avós
mais amorosos de que há memória! Obrigada por me acompanharem desde
sempre e por acreditarem em mim como mais ninguém acredita.
Aos meus queridos pais. Estou muito grata por tudo o que me ensinaram e
transmitiram, por todas as oportunidades que me proporcionaram, e por todo o
amor e carinho que me dão sempre. Nunca vos poderei agradecer o suficiente.
E, por fim, ao Zinho, pelo amor, amizade e compreensão, e pelo apoio

incondicional em todas as etapas. Temos ainda muitas pela frente, mas juntos
chegaremos onde queremos! Por tudo isso e muito mais, sei que tenho uma
sorte tremenda por te ter na minha vida.
palavras-chave Microrganismos, fermentação, glicerol, stress, adaptação, alta pressão.
resumo A conversão microbiana de glicerol em compostos de valor-acrescentado

constitui uma aplicação promissora para o excesso de glicerol gerado pela
indústria do biodiesel. Uma possível estratégia para melhorar estes processos
passa pelo uso de níveis sub-letais de Alta Pressão (AP) ao longo da
fermentação, de modo a estimular o crescimento dos microrganismos
fermentativos e/ou aumentar a velocidade e os rendimentos do processo.
Assim, o presente trabalho teve como objetivo o estudo da aplicação de AP
(entre 10 e 50 MPa) em dois processos fermentativos: um deles com a
Lactobacillus reuteri, que produz 1,3-propanediol a partir de glicerol; e outro
com a Paracoccus denitrificans, que produz polihidroxialcanoatos a partir de
glicerol.
No caso da fermentação com a L. reuteri, a produção de 1,3-propanediol foi
estimulada a 10 MPa, resultando em aumentos de rendimento e produtividade
de ≈ 11 e 12 %, respetivamente, relativamente às amostras à pressão
atmosférica. A adaptação da L. reuteri à pressão foi ainda avaliada durante
quatro ciclos de fermentação consecutivos. Após o quarto ciclo a 10 MPa, a
concentração de 1,3-propanediol aumentou 15 % relativamente ao respetivo
ciclo à pressão atmosférica, e ainda mais (52 %) relativamente à “abordagem
convencional”, i.e. sem os ciclos de fermentação, à pressão atmosférica. Estes
resultados confirmam que diferentes estratégias de aplicação de AP podem
aumentar a produção de 1,3-propanediol. De modo a compreender os efeitos
dos ciclos de pressão no metabolismo da L. reuteri, realizou-se um estudo
metabolómico comparativo entre amostras de fermentação a diferentes
condições. Foram observados perfis metabólicos distintos de acordo com a
pressão aplicada e com o ciclo fermentativo, sendo que o efeito adaptativo ao
longo dos ciclos foi mais acentuado a 10 MPa. Estes resultados fornecem
informação relevante relativa à adaptação de bactérias ácido-láticas a
pressões sub-letais.
Numa fase inicial do estudo da fermentação de glicerol pela P. denitrificans,
analisou-se o efeito da disponibilidade de ar no processo, uma vez que este
corresponde a um parâmetro crítico que afeta o crescimento celular e o perfil
metabólico deste microrganismo, e poderá ser limitado durante a fermentação
sob pressão. De facto, amostras com elevada disponibilidade de ar mostraram
crescimento celular considerável, mas ausência de produção de etanol ou
ácidos orgânicos. Por outro lado, amostras sem ar mostraram reduzido
crescimento celular, mas atividade metabólica ativa, com produção de etanol,
succinato e acetato. Relativamente ao estudo do efeito da pressão, a P.
denitrificans mostrou capacidade para crescer a 10, 25 e 35 MPa, ainda que
não tão acentuadamente quanto à pressão atmosférica. O uso da AP
promoveu alterações nos perfis fermentativos da P. denitrificans. Para além
disso, a AP afetou a produção de polihidroxialcanoatos, resultando em
concentrações de polímero mais baixas, mas conteúdos de polímero em
biomassa (%) superiores, o que indica uma maior capacidade de acumular
este composto no interior das células. Sob determinados níveis de pressão, os
polímeros formados apresentaram modificações na sua composição
monomérica, relativamente aos obtidos à pressão atmosférica. É possível
antever que estas alterações poderão afetar as propriedades físicas e
mecânicas do polímero.
Em suma, o presente trabalho revelou diferentes caraterísticas metabólicas da
L. reuteri e da P. denitrificans. Os resultados obtidos mostraram ainda que,
quando aplicada em condições adequadas, a tecnologia de AP poderá
apresentar-se como uma ferramenta útil para modificar e até melhorar
processos de fermentação de glicerol, bem como os respetivos produtos. Este
estudo abre também os horizontes para o uso desta tecnologia no estímulo de
outros processos fermentativos que sejam relevantes para a indústria
biotecnológica.
keywords Microorganisms, fermentation, glycerol, stress, adaptation, high pressure.
abstract Microbial conversion of glycerol into value-added products constitutes a

promising approach to dispose the excess glycerol generated by the biodiesel
industry. A possible strategy to improve these processes is the performance of
fermentations under sub-lethal high pressure (HP), which may stimulate cell
growth and/or increase fermentation rates and yields. Therefore, the goal of the
present work was to study the application of HP (in the range of 10 to 50 MPa)
on two different microbial processes: one of them with Lactobacillus reuteri,
which is able to produce 1,3-propanediol from glycerol; and the other with
Paracoccus denitrificans, which is able to produce polyhydroxyalkanoates from
glycerol.
In the case of L. reuteri fermentation, the production of 1,3-propanediol was
stimulated at 10 MPa, resulting in yield and productivity improvements of ≈ 11
and 12 %, respectively, relatively to the same samples at atmospheric
pressure. Adaptation of L. reuteri to pressure was assessed during four
consecutive fermentation cycles. After the fourth cycle at 10 MPa, 1,3-
propanediol titers increased 15 % relative to the respective cycle at
atmospheric pressure, and even more (52 %) relative to the “conventional
approach”, i.e. without the fermentation cycles, at atmospheric pressure. These
results confirm that different strategies of HP application may stimulate the
production of 1,3-propanediol. To better understand the effects of these HP-
cycles on the L. reuteri metabolism, a comparative metabolomic study between
L. reuteri fermentation samples was performed. The results showed distinct
metabolic profiles according to the pressure applied and the fermentation
cycles. The adaptive effect throughout the cycles was considerably more
accentuated at 10 MPa. These results unveil relevant information regarding the
adaptation of lactic acid bacteria to sub-lethal HP.
Regarding glycerol fermentation by P. denitrificans, it was necessary to give
special focus on air availability, since this is a critical parameter that affects cell
growth and metabolic profile, and it may be limited during fermentation under
HP conditions. Samples with higher air availability showed considerable cell
growth, but no production of ethanol, acetate and succinate. On the other hand,
samples without air had lower cell growth, but active metabolic activity (with the
production of organic acids and ethanol). Regarding the experiments of
fermentation under HP, P. denitrificans was able to grow at 10, 25 and 35 MPa,
but at lower extent compared to atmospheric pressure. Application of HP
promoted modifications in the P. denitrificans fermentative profiles at different
pressure levels. In addition, HP was found to affect polyhydroxyalkanoate
production, resulting in lower titers, but higher polymer content in cell dry mass
(%), which indicates higher ability to accumulate these polymers in the cells.
Some levels of HP also affected PHA monomeric composition, showing
considerable differences relative to the ones obtained at atmospheric pressure.
It is possible to foresee that these changes in polymer composition may also
affect its physical and mechanical properties.
Overall, the present work demonstrated new metabolic features of L. reuteri
and P. denitrificans. The obtained results showed that HP technology (applied
at specific levels) can be a useful tool to modify and possible improve glycerol
fermentation processes and products. It also opens the possibility of using this
technology to stimulate other fermentations relevant for the biotechnological
field.
TABLE OF CONTENTS
List of Figures ............................................................................................................................... i

List of Tables ............................................................................................................................... v
List of Publications ....................................................................................................................vii
List of Abbreviations ................................................................................................................... xi
Chapter I. Fermentation at non-conventional conditions and its application to food- and

bio-sciences: A literature revision ............................................................................................. 1
1.1. General overview .............................................................................................................. 3
1.2. Fermentation under high pressure conditions .................................................................... 4
1.2.1. Technological equipment suitable for fermentation processes .................................. 8
1.3. Comparison between high pressure and other non-conventional technologies .................. 9
1.4. References ....................................................................................................................... 12
Chapter II. Fermentation of glycerol for production of valuable bio-chemicals: A

literature revision ..................................................................................................................... 17
2.1. The biodiesel industry and the glycerol surplus ............................................................... 19
2.2. Glycerol bio-conversion into valuable products .............................................................. 20
2.3. Glycerol bio-conversion into 1,3-propanediol ................................................................. 23
2.3.1. Production of 1,3-propanediol by Lactobacillus reuteri ........................................... 25
2.4. Glycerol bio-conversion into polyhydroxyalkanoates ..................................................... 29
2.4.1. Polyhydroxyalkanoate production by Paracoccus denitrificans ............................... 30
2.5. Current challenges and possible improvements ............................................................... 32
2.6. References ....................................................................................................................... 34
Chapter III. Scope and outline of the work ............................................................................ 45
Chapter IV. Lactobacillus reuteri growth and fermentation under high pressure towards
the production of 1,3-propanediol........................................................................................... 49
4.1. Introduction ..................................................................................................................... 51
4.2. Material and methods ...................................................................................................... 52
4.2.1. Microorganism and culture media ............................................................................ 52
4.2.2. Seed culture preparation ........................................................................................... 53
4.2.3. Fermentation experiments ........................................................................................ 53
4.2.4. Determination of biomass concentration .................................................................. 53
4.2.5. Determination of viable cell counts .......................................................................... 54
4.2.6. Quantification of glycerol, glucose, and fermentation products ............................... 54
4.2.7. Statistical analysis .................................................................................................... 54
4.3. Results and discussion ..................................................................................................... 54
4.3.1. Effect of sub-lethal pressure on cell growth ............................................................. 54
4.3.2. Effect of sub-lethal pressure on substrate consumption ............................................ 57
4.3.3. Effect of sub-lethal pressure on production of 1,3-propanediol ................................ 59
4.3.4. Effect of sub-lethal pressure on formation of fermentation by-products................... 62
4.4. Conclusions ..................................................................................................................... 66
4.5. References ....................................................................................................................... 68
Chapter V. Utilization of glycerol during consecutive cycles of Lactobacillus reuteri

fermentation under pressure: The impact on cell growth and fermentation profile .......... 73
5.1. Introduction ..................................................................................................................... 75
5.2. Material and methods ...................................................................................................... 76
5.2.1. Microorganism and culture media ............................................................................ 76
5.2.2. Seed culture preparation ........................................................................................... 76
5.2.3. Fermentation experiments ........................................................................................ 77
5.2.4. Determination of biomass concentration .................................................................. 78
5.2.5. Determination of viable cell counts .......................................................................... 78
5.2.6. Quantification of glycerol, glucose, and fermentation products ............................... 78
5.2.7. Analysis of the leakage of nucleic acids and proteins ............................................... 78
5.2.8. Statistical analysis .................................................................................................... 79
5.3. Results and discussion ..................................................................................................... 79
5.3.1. Cell growth ............................................................................................................... 79
5.3.2. Substrate consumption ............................................................................................. 81
5.3.3. Production of 1,3-propanediol .................................................................................. 83
5.3.4. Production of acetate, lactate and ethanol................................................................. 87
5.3.5. Leakage of proteins and nucleic acids ...................................................................... 91
5.4. Conclusions ..................................................................................................................... 94
5.5. References ....................................................................................................................... 96
Chapter VI. Comparative metabolomic profiling of Lactobacillus reuteri under high

pressure fermentation cycles ................................................................................................. 101
6.1. Introduction ................................................................................................................... 103
6.2. Material and methods .................................................................................................... 105
6.2.1. Fermentation experiments ...................................................................................... 105
6.2.2. 1H NMR experiments ............................................................................................. 106
6.2.3. Multivariate data analysis ....................................................................................... 106
6.3. Results and discussion ................................................................................................... 107
6.4. Conclusions ................................................................................................................... 115
6.5. References ..................................................................................................................... 117
Chapter VII. The use of different fermentative approaches on Paracoccus denitrificans:

Effect of high pressure and air availability on growth and metabolism ............................ 119
7.1. Introduction ................................................................................................................... 121
7.2.1. Microorganism and culture media .......................................................................... 122
7.2.2. Inoculum preparation and inoculation .................................................................... 123
7.2.3. Fermentation experiments: Effect of air availability............................................... 123
7.2.4. Fermentation experiments: Effect of high pressure ................................................ 123
7.2.5. Analysis of biomass concentration ......................................................................... 124
7.2.6. Glycerol quantification ........................................................................................... 124
7.2.7. Characterization of the extracellular medium ......................................................... 124
7.2.8. Statistical analysis .................................................................................................. 124
7.3.1. Effect of air availability on Paracoccus denitrificans growth and fermentation at
atmospheric pressure ........................................................................................................ 125
7.3.2. Effect of high pressure on Paracoccus denitrificans growth and fermentation ..... 129
7.4. Conclusions ................................................................................................................... 135
7.5. References ..................................................................................................................... 137
Chapter VIII. Effect of high pressure on Paracoccus denitrificans growth and

polyhydroxyalkanoates production from glycerol .............................................................. 141
8.1. Introduction ................................................................................................................... 143
8.2.1. Microorganism and culture media .......................................................................... 145
8.2.2. Seed culture preparation ......................................................................................... 145
8.2.3. Fermentation experiments ...................................................................................... 145
8.2.4. Determination of biomass concentration ................................................................ 146
8.2.5. Determination of viable cell counts ........................................................................ 146
8.2.6. Glycerol quantification ........................................................................................... 146
8.2.7. Analysis of polyhydroxyalkanoate formation ......................................................... 146
8.2.8. Statistical analysis .................................................................................................. 147
8.4. Conclusions ................................................................................................................... 155
8.5. References ..................................................................................................................... 157
Chapter IX. Conclusions and outlook................................................................................... 161
Appendix A. Additional information supporting Chapter VI ........................................... 167
Appendix B. Additional information supporting Chapter VII .......................................... 173

LIST OF FIGURES
Chapter I
Figure 1.1. Schematic representation of microbial fermentation under high pressure used at
laboratory scale ............................................................................................................................ 9
Chapter II
Figure 2.1. Metabolic pathway of glycerol bioconversion: oxidative and reductive wings of
glycerol assimilation .................................................................................................................. 22
Figure 2.2. Lactobacillus reuteri metabolic pathways from glucose and glycerol ..................... 26
Chapter IV
Figure 4.1. Biomass concentration and viable cell counts throughout fermentation, for samples
with initial acetate (a and b) and without initial acetate (c and d), according to the pressure applied:
10 MPa, 25 MPa, or 35 MPa. Control samples (0.1 MPa) are represented as squares ................ 55
Figure 4.2. Glycerol and glucose concentrations throughout fermentation, for samples with initial
acetate (a and b) and without initial acetate (c and d), according to the pressure applied: 10 MPa,
25 MPa, or 35 MPa. Control samples (0.1 MPa) are represented as squares .............................. 57
Figure 4.3. 1,3-Propanediol concentration throughout fermentation, for samples with initial
acetate (a) and without initial acetate (b), according to the pressure applied: 10 MPa, 25 MPa, or
35 MPa. Control samples (0.1 MPa) are represented as squares ................................................ 59
Figure 4.4. Lactate, acetate and ethanol concentrations throughout fermentation, for samples with
initial acetate (a, b and c) and without initial acetate (c, d and e), according to the pressure applied:
10 MPa, 25 MPa, or 35 MPa. Control samples (0.1 MPa) are represented as squares ................ 63
Chapter V
Figure 5.1. Schematic representation of the fermentation experiments, carried out in four
consecutive cycles of 24 h, at 10 or 25 MPa. Fermentation cycles were also performed at 0.1 MPa,
to use as control.......................................................................................................................... 77
Figure 5.2. Biomass concentration and viable cell counts after each fermentation cycle (24 h), for
samples with an initial acetate concentration of 5 g L-1 (a and b), and for samples without initial
acetate (c and d), according to the pressure applied: 0.1 MPa (control samples), 10 MPa, and 25
MPa ............................................................................................................................................ 80
i
Figure 5.3. Glycerol and glucose concentrations after each fermentation cycle (24 h), for samples
with an initial acetate concentration of 5 g L -1 (a and b), and for samples without initial acetate (c
and d), according to the pressure applied: 0.1 MPa (control samples), 10 MPa, and 25 MPa .... 82
Figure 5.4. 1,3-Propanediol concentration after each fermentation cycle (24 h), for samples with
an initial acetate concentration of 5 g L-1 (a), and for samples without initial acetate (b), according
to the pressure applied: 0.1 MPa (control samples), 10 MPa, and 25 MPa ................................. 84
Figure 5.5. Lactate, acetate and ethanol concentrations after each fermentation cycle (24 h), for
samples with an initial acetate concentration of 5 g L-1 (a, b and c), and for samples without initial
acetate (d, e and f), according to the pressure applied: 0.1 MPa (control samples), 10 MPa, and
25 MPa ....................................................................................................................................... 88
Figure 5.6. Ratio of extracellular nucleic acid and protein contents (%) after each fermentation
cycle (24 h) at 10 and 25 MPa, relative to 0.1 MPa, for samples with an initial acetate
concentration of 5 g L-1 (a and b), and for samples without initial acetate (c and d) .................. 92
Chapter VI
Figure 6.1. PCA scores plot of L. reuteri fermentation metabolites, obtained by 1D 1H NMR. C1
and C4 samples correspond to the first and fourth fermentation cycles at atmospheric pressure,
respectively; P1 and P4 samples correspond to the first and fourth fermentation cycles at 10 MPa,
respectively. In all cases, the results are presented in duplicated (indicated as 1 and 2) ........... 109
Figure 6.2. Heat map of L. reuteri metabolites at different fermentation cycles and pressure
conditions. Scale is based on colors from red to blue representing an increase and a decrease in
metabolite levels, respectively. The C1 and C4 samples correspond to the first and fourth
fermentation cycles at atmospheric pressure, respectively; P1 and P4 samples correspond to the
first and fourth fermentation cycles at 10 MPa, respectively. The signals are indicated by a code
name, between M1 and M73 .................................................................................................... 110
Figure 6.3. Volcano plots showing the differences in the metabolic profiles of L. reuteri between
the first and the fourth fermentation cycles, at 0.1 MPa (a) and at 10 MPa (b). The
signals/metabolites are indicated by a code name, between M1 and M73 ................................ 112
Figure 6.4. Volcano plots showing the differences between the metabolic profiles of L. reuteri at
10 and 0.1 MPa, after the fourth fermentation cycle. The signals/metabolites are indicated by a
code name, between M1 and M73 ............................................................................................ 113
Figure 6.5. Schematic representation of general metabolic pathways for 2,3-butanediol

production by glucose metabolism, in lactic acid bacteria ....................................................... 114
ii
Chapter VII
Figure 7.1. Concentrations of biomass (a) and glycerol (b) throughout fermentation, for different
air availability conditions: without air during the entire process; with air during the first 24 h, and
without air during the remaining 48 h; or with air during the entire process ............................ 125
Figure 7.2. Concentrations of ethanol (a), acetate (b) and succinate (c) throughout fermentation,
for different air availability conditions: without air during the entire process; with air during the
first 24 h, and without air during the remaining 48 h; or with air during the entire process ..... 126
Figure 7.3. Concentrations of biomass (a) and glycerol (b) throughout fermentation at different
HP conditions: 10 MPa, 25 MPa, or 35 MPa. Control samples (0.1 MPa) are also represented
................................................................................................................................................. 129
Figure 7.4. Glucose concentrations throughout fermentation at different HP conditions: 10 MPa,

25 MPa, or 35 MPa. Control samples (0.1 MPa) are also represented ...................................... 131
Figure 7.5. Concentrations of ethanol (a), acetate (b) and succinate (c) throughout fermentation
at different HP conditions: 10 MPa, 25 MPa, or 35 MPa. Control samples (0.1 MPa) are also
represented .............................................................................................................................. 132
Chapter VIII
Figure 8.1. Biomass concentration over time and viable cell counts after 72 h, for fermentation
at different pressure conditions: 10, 25, 35 and 50 MPa. Control samples, corresponding to
fermentation at 0.1 MPa, are also indicated as C1 (higher air availability and agitation) and C2
(lower air availability and no agitation).................................................................................... 148
Figure 8.2. Glycerol concentrations over time, for fermentation at different pressure conditions:
10, 25, 35 and 50 MPa. Control samples, corresponding to fermentation at 0.1 MPa, are also
indicated as C1 (higher air availability and agitation) and C2 (lower air availability and no
agitation) .................................................................................................................................. 149
Appendix A
Figure A.1. Examples of 1H NMR spectra of fermentation samples after one and four cycles at
atmospheric pressure (C1 and C4, respectively), and at 10 MPa (P1 and P4, respectively): (a) full
spectra; and expansions for (b) aromatic region (6.0-9.0 ppm) and (c) aliphatic region (0.6-3.0
ppm) ......................................................................................................................................... 169
Figure A.2. Principal component analysis (PCA) loading plots 1 (a) and 2 (b) of L. reuteri
fermentation metabolites, obtained by 1D 1H NMR ................................................................. 170
iii
Figure A.3. Metabolite plots showing the abundance of 1,3-PDO (for M44 and M60, both signals
corresponding to this metabolite) in samples at 0.1 MPa (C1 and C4, for the first and fourth
fermentation cycles, respectively) and at 10 MPa (P1 and P4 samples, for the first and fourth
fermentation cycles, respectively) ............................................................................................ 171
iv
LIST OF TABLES
Chapter I
Table 1.1. General effects of high pressure on microbial cell growth and fermentation .............. 6
Chapter II
Table 2.1. Some of the most relevant bioproducts obtained from glycerol in fermentation by
different microorganisms .......................................................................................................... 20
Table 2.2. Comparison of maximum theoretical yield for production of different compounds
from glycerol and glucose .......................................................................................................... 21
Table 2.3. Main results of 1,3-PDO production of by L. reuteri, using glycerol as co-substrate….
.................................................................................................................................................. 27
Table 2.4. Main results of PHA production by P. denitrificans, using glycerol as substrate ..... 31
Chapter IV
Table 4.1. Consumption (%) of glycerol and glucose after 32 h of fermentation at different
pressure conditions, in samples with and without acetate .......................................................... 58
Table 4.2. 1,3-PDO yields on glycerol (Y1,3-PDO/Gly) and 1,3-PDO productivities (Q1,3-PDO), for
fermentation at different pressure conditions, in samples with and without acetate .................. 60
Table 4.3. Yields of lactate (YLact/Glu), acetate (YAcet/Glu), and ethanol (YEtOH/Glu) on glucose, for
fermentation at different pressure conditions, in samples with and without acetate .................. 64
Table 4.4. Molar ratios between 1,3-PDO and by-products (lactate, acetate, and ethanol) produced
after 32 h of fermentation at different pressure conditions, in samples with and without acetate
................................................................................................................................................... 65
Chapter V
Table 5.1. 1,3-Propanediol yields on glycerol (Y1,3-PDO/Gly) and 1,3-PDO productivities (Q1,3-PDO),
at the end of each fermentation cycle (24 h), at different pressure conditions, in samples with and
without acetate ........................................................................................................................... 85
Table 5.2. Yields of lactate (YLact/Glu), acetate (YAcet/Glu), and ethanol (YEtOH/Glu) on glucose, at the
end of each fermentation cycle (24 h), at different pressure conditions, in samples with and
without acetate ........................................................................................................................... 89
v
Table 5.3. Molar ratios between 1,3-PDO and by-products (lactate, acetate, and ethanol) produced
after each fermentation cycle (24 h), at different pressure conditions, in samples with and without
acetate ........................................................................................................................................ 91
Chapter VI
Table 6.1. Comparison between different metabolomics technologies (NMR spectroscopy, GC-
Mass spectrometry, and LC-Mass spectrometry) ..................................................................... 104
Table 6.2. List of relevant compounds identified in the samples by comparison with databases
and an appropriate software, with the respective chemical shifts and code names ................... 107
Chapter VII
Table 7.1. Yields of biomass (YX/S), ethanol (YEtOH/S), acetate (YAcet/S), and succinate (YSucc/S) on
glycerol, for fermentation under different air availability conditions ....................................... 127
Table 7.2. Specific growth rates (µ), final biomass concentrations and percentages of glycerol
consumed, after 24 h of fermentation, for medium:air ratios of 1.0:1.8 and 1.0:2.2 ................. 129
Table 7.3. Yields of biomass (YX/S), ethanol (YEtOH/S), acetate (YAcet/S), and succinate (YSucc/Gly) on
glycerol, for fermentation under different pressure conditions ................................................. 134
Chapter VIII
Table 8.1. Specific growth rates (µ), percentages of glycerol consumed, and yields of biomass on
glycerol, after 72 h of fermentation, at different pressure conditions (0.1 – 35 MPa)............... 149
Table 8.2. Polyhydroxyalkanoate contents, yields and productivities, after 72 h of fermentation,

at different pressure conditions (0.1 – 35 MPa). ....................................................................... 151
Table 8.3. Polyhydroxyalkanoate monomeric composition (mol%), after 72 h of fermentation, at

different pressure conditions (0.1 – 35 MPa). .......................................................................... 153
Appendix B
Table B.1. Glucose and maltose concentrations, at 0 h and 72 h, for fermentation under different
air availability conditions ......................................................................................................... 175
vi
LIST OF PUBLICATIONS
This thesis is based on the following scientific articles:
Fermentation at non-conventional conditions in food- and bio-sciences by application of

advanced processing technologies
Mota, M. J.; Lopes, R. P.; Koubaa, M.; Roohinejad, S.; Barba, F. J.; Delgadillo, I.; Saraiva, J. A.,
Critical Reviews in Biotechnology 2018, 38 (1), 122-140 (doi:
10.1080/07388551.2017.1312272).
Lactobacillus reuteri growth and fermentation under high pressure towards the production
of 1,3-propanediol
Mota, M.J., Lopes, R.P., Sousa, S., Gomes, A.M., Delgadillo, I., Saraiva, J.A., Submitted in Food
Research International.
Utilization of glycerol during consecutive cycles of Lactobacillus reuteri fermentation under

pressure: The impact on cell growth and fermentation profile
Mota, M.J., Lopes, R.P., Sousa, S., Gomes, A.M., Lorenzo, J.M., Barba, F.J., Delgadillo, I.,
Saraiva, J.A., Submitted in Process Biochemistry.
The use of different fermentative approaches on Paracoccus denitrificans: Effect of high

pressure and air availability on growth and metabolism
Mota, M.J., Lopes, R.P., Pinto, C.A, Sousa, S., Gomes, A.M., Delgadillo, I., Saraiva, J.A.,
Submitted in Biotechnology Progress.
Effect of high pressure on Paracoccus denitrificans growth and polyhydroxyalkanoates

production from glycerol
Mota, M.J., Lopes, R.P., Simões, M.M.Q., Delgadillo, I., Saraiva, J.A., Submitted in Journal of
Industrial Microbiology & Biotechnology.
vii
During my studies, I have also contributed to the following scientific articles that are not included
in this thesis:
Effect of ultrasound on lactic acid production by Lactobacillus strains in date (Phoenix

dactylifera var. Kabkab) syrup
Hashemi, S. M. B.; Khaneghah, A. M.; Saraiva, J. A.; Jambrak, A. R.; Barba, F. J.; Mota, M. J.,
Applied Microbiology and Biotechnology 2018, 102 (6), 2635–2644 (doi: 10.1007/s00253-018-
8789-8).
Application of High Pressure with Homogenization, Temperature, Carbon Dioxide, and

Cold Plasma for the Inactivation of Bacterial Spores: A Review
Lopes, R. P.; Mota, M. J.; Gomes, A. M.; Delgadillo, I.; Saraiva, J. A., Comprehensive Reviews
in Food Science and Food Safety 2018, 17 (3), 532-555 (doi: 10.1111/1541-4337.12311).
Growth and metabolism of Oenococcus oeni for malolactic fermentation under pressure
Neto, R.; Mota, M. J.; Lopes, R. P.; Delgadillo, I.; Saraiva, J. A., Letters in Applied Microbiology
2016, 63 (6), 426-433 (doi:10.1111/lam.12664).
Probiotic yogurt production under high pressure and the possible use of pressure as an
on/off switch to stop/start fermentation
Mota, M. J.; Lopes, R. P.; Delgadillo, I.; Saraiva, J. A., Process Biochemistry 2015, 50 (6), 906-
911 (doi:10.1016/j.procbio.2015.03.016).
Fruit juice sonication: Implications on food safety and physicochemical and nutritional
properties
Zinoviadou, K. G.; Galanakis, C. M.; Brnčić, M.; Grimi, N.; Boussetta, N.; Mota, M. J.; Saraiva,
J. A.; Patras, A.; Tiwari, B.; Barba, F. J., Food Research International 2015, 77 (4), 743-752
(doi:10.1016/j.foodres.2015.05.032).
Preservation of a highly perishable food, watermelon juice, at and above room temperature
under mild pressure (hyperbaric storage) as an alternative to refrigeration
Santos, M. D.; Queirós, R. P.; Fidalgo, L. G.; Inácio, R. S.; Lopes, R. P.; Mota, M. J.; Sousa, S.
G.; Delgadillo, I.; Saraiva, J. A., LWT – Food Science and Technology 2015, 62 (1), 901-905
(doi:10.1016/j.lwt.2014.06.055).
viii
Hyperbaric storage of melon juice at and above room temperature and comparison with
storage at atmospheric pressure and refrigeration
Queirós, R. P.: Santos, M. D.; Fidalgo, L. G.; Mota, M. J.; Lopes, R. P.; Inácio, R. S.; Delgadillo,
I.; Saraiva, J. A., Food Chemistry 2014, 147, 209-214 (doi:10.1016/j.foodchem.2013.09.124).
Hyperbaric storage at and above room temperature of a highly perishable food

Fidalgo, L. G.; Santos, M. D.; Queirós, R. P.; Inácio, R. S.; Mota, M. J.; Lopes, R. P.; Gonçalves,
M. S.; Neto, R. F.; Saraiva, J. A., Food Bioprocess Technology 2014, 7 (7), 2028-2037
(doi:10.1007/s11947-013-1201-x).
ix
x
LIST OF ABBREVIATIONS
1,3-PDO 1,3-propanediol
2,3-BDO 2,3-butanediol
3-HP 3-hydroxypropinonic acid
3-HPA 3-hydroxypropionaldehyde
ATP adenosine triphosphate
CDW cell dry weight
DHA dihydroxyacetone
DNA deoxyribonucleic acid
EMP Embden–Meyerhof–Parnas
FDR false discovery rate
FID free induction decay
GC gas chromatography
GRAS generally recognized as safe
HP high pressure
HPLC high-performance liquid chromatography
HSPs heat-shock proteins
LAB lactic acid bacteria
LC liquid chromatography
LDH lactate dehydrogenase
mcl-PHA medium chain length polyhydroxyalkanoates
MEF moderate electric fields
MS mass spectrometry
NADH/NAD+ nicotinamide adenine dinucleotide
NADPH/NADP+ nicotinamide adenine dinucleotide phosphate
MRS de Man, Rogosa and Sharpe
NMR nuclear magnetic resonance
OH ohmic heating
PCA principal component analysis
pdu propanediol-utilization operon
PEF pulsed electric fields
PEP phosphoenolpyruvate
PHA polyhydroxyalkanoates
PLS-DA partial least squares discriminant analysis
xi
PTT polytrimethylene terephthalate
RI refraction index
scl-PHA short chain length polyhydroxyalkanoates
TCA tricarboxylic acid
tmRNA transfer-messenger ribonucleic acid
TSP-d4 3-(trimethylsilyl)propionic-2,2,3,3-d4 acid
US ultrasound
UV ultraviolet
xii
CHAPTER I
Fermentation at non-conventional conditions

and its application to food- and bio-sciences:
A literature revision
Adapted from:
Mota, M.J., Lopes, R.P., Koubaa, M., Roohinejad, S., Barba, F.J.; Delgadillo, I., Saraiva, J.A.,
2018. Fermentation at non-conventional conditions in food- and bio-sciences by application of
advanced processing technologies. Crit. Rev. Biotechnol. 38, 122–140.
CHAPTER I. Fermentation at non-conventional conditions and its application to food- and bio-sciences
1.1. General overview

Microorganisms may show different reactions when exposed to stressful conditions. At
more extreme stress conditions, cells are unable to withstand and adapt, leading to microbial
destruction. Food pasteurization is based on this principle, since the main goal is to destroy
vegetative microbial contaminants in order to assure food safety and further preservation. In
contrast, when mild stress conditions are applied (at sub-lethal levels), the microorganisms may
be able to survive due to the activation of specific stress response mechanisms, as well as possible
adaptation to the new conditions (Huang et al., 2014; Lado and Yousef, 2002). Stress sensing
systems and defenses are developed by microorganisms to allow them to withstand extreme
conditions and sudden environmental changes. Bacterial stress responses rely on the coordinated
expression of genes that alter different cellular processes (cell division, DNA metabolism,
housekeeping, membrane composition, transport, etc.) to improve the stress tolerance (Storz and
Hengge, 2000; van de Guchte et al., 2002). As a consequence, several changes in microbial
morphology and metabolic reactions are also observed (Huang et al., 2014). Some of the stress-
induced genes seem to be genuinely specific, while others are induced by a wide variety of stresses
and are thought to be general stress response genes (De Angelis et al., 2001; Serrazanetti et al.,
2009). Responses are different and vast and depend on the species, strain, and the type of stress
applied (Mills et al., 2011; Serrazanetti et al., 2009; van de Guchte et al., 2002; Zamfir and Grosu-
Tudor, 2014).
The implications and the involvement of the metabolic processes in the stress responses
could affect fermentations, possibly resulting in improvements of bioprocesses and bioproducts
relevant for different industries (Serrazanetti et al., 2009). As a result, the concept of performing
fermentations at non-conventional conditions is arising, and it is based on the use of emerging
processing technologies (typically applied for food pasteurization). Pulsed electric fields (PEF),
moderate electric fields (MEF), ohmic heating (OH), ultrasound (US) and high pressure (HP) are
therefore applied at sub-lethal levels, at some point during the fermentative processes, affecting
the growth and metabolism of the microbial strains involved in fermentation, but without causing
microbial inactivation.
Several microorganisms were tested under non-conventional conditions, including both
bacteria and yeasts, and ultrasound was the most studied technology for this purpose. Considering
the promising results obtained so far, we foresee that this subject will become a research trend
within the next few years in microbial biotechnology. We believe that there is a great potential to
explore, since several fermentative processes with a prominent role in different industries could
benefit from this approach. These industrially relevant processes include not only food
fermentations (e.g. for the production of dairy products, alcoholic beverages, and others), but also
3
the production of commodity biochemicals (such as acetate, citrate, and ethanol) and high-value
bioproducts (such as vitamins, antibiotics, and biopolymers).
The following section focused on the application of HP to different fermentation
processes, while providing a perspective regarding the strengths and limitations of the HP
equipment currently available for fermentation processes. The use of electric fields and US in
fermentation was already reviewed by Galván-D’Alessandro and Carciochi (2018), and it was not
addressed in detail in the present work, since they were not studied here.
1.2. Fermentation under high pressure conditions

High pressure is an emerging technology with increasingly successful industrial
application as a non-thermal food pasteurization method (Barba et al., 2015b, 2012). The
mechanisms of HP on microbial cells are already well understood, with identification of several
effects on metabolism, physiology and structural organization (Bartlett, 2002). With the
increasing pressure, relevant cell structures and functions are successively compromised until it
turns impossible to withstand the stress and survive at these hostile conditions (Mota et al., 2013).
In terms of cell structure, different organelles show different sensitivity to high pressure. For
instance, lipid membranes are particularly pressure sensitive, because of its high compressible
potential. Thus, changes in membrane composition and fluidity are observed under HP, as well
as weakening of important protein-lipid interactions (Winter and Jeworrek, 2009). High pressure
treatments may also affect the structure of DNA, ribosomes and proteins (Abe, 2007; Macgregor
Jr, 2002; Niven et al., 1999), possibly leading to inhibition of cell processes (such as replication,
transcription and translation) and metabolic reactions essential for cell maintenance.
The magnitude of cell damage by HP is highly dependent on several parameters, which
include the level and duration of the pressure treatments, the compression method and other
environmental conditions (temperature, media composition, pH, etc.). In addition, each microbial
strain has a specific degree of HP tolerance according to their intrinsic cellular characteristics. In
general, prokaryotes are more HP-resistant than eukaryotes, Gram-positive bacteria are more HP-
resistant Gram-negative bacteria, and cocci are more HP-resistant than bacilli (Huang et al.,
2014). The cell growth stage was also found to affect the microbial tolerance to HP treatments,
which is usually higher during the stationary phase than during the exponential phase. This can
be explained by the lower stress tolerance of cells during the exponential phase, due to the
continuous cell division and synthesis. In contrast, microorganisms in the stationary phase have
complete cell structures, thus they are able to resist more severe stress levels (Huang et al., 2014;
Patterson, 2005). Moreover, Hill et al. (2002) reported that the higher pressure resistance observed
4
during the stationary-phase is partially due to the synthesis of proteins that protect against a range
of adverse conditions.
It is important to note that microorganisms are more likely to be stressed or injured than
killed under HP, particularly when lower intensity treatments are applied (Huang et al., 2014).
Several studies have found that microorganisms possess regulatory genes for environmental
adaptation, generally involving the accumulation of heat-shock proteins (HSPs) within the cell to
enhance the resistance to multiple environmental stresses (Lou and Yousef, 1997; Wemekamp-
Kamphuis et al., 2004). Welch et al. (1993) reported that a HP treatment of 55 MPa induced a
stress response by production of cold-shock proteins, heat-shock proteins, and other protective
proteins. Hörmann et al. (2006) used a comparative proteome approach to characterize the HP
effects on Lactobacillus sanfranciscensis, concluding that HP stress response uses subsets of
other stress responses (such as cold and high salinity).
Application of sub-lethal HP treatments leads to possible acquisition of new desirable
characteristics, obtained by inhibition or even suppression of some metabolic pathways and/or
utilization of new ones (Mota et al., 2013). This concept is gaining relevance over the last years,
since the piezo-tolerant strains may have numerous interesting applications in different fields
(Aertsen et al., 2009; Hörmann et al., 2006). The main studies regarding HP application of
microbial fermentation processes are indicated in Table 1.1.
Most studies regarding the effects of HP on microbial growth and metabolism were performed
with Saccharomyces cerevisiae. Picard et al. (2007) observed that at 10 MPa, the fermentation of
glucose to ethanol by S. cerevisiae proceeded 3-fold faster than control fermentation. In addition,
at 5 and 10 MPa the fermentation yield was enhanced by 6 % and 5 %, respectively, compared to
the control. However, at pressures above 20 MPa fermentation was slowed down, and the authors
estimated that alcoholic fermentation was interrupted at 87 ± 7 MPa, possibly due to loss of
activity by one or more enzymes involved in the glycolytic pathway. Similarly, Bravim et al.
(Bravim et al., 2012) observed that pre-treatment of S. cerevisiae with HP led to an increase in
ethanol content upon fermentation. A global transcriptional analysis revealed the over-expression
of several genes related to cell recovery and stress tolerance induced by HP. The most relevant
case was the gene SYM1, which was related to enhancement of ethanol production and increase
of stress tolerance upon fermentation.
Trehalose and glutathione are two major stress-induced metabolites with industrial value,
which could be produced under HP conditions. S. cerevisiae CICC1339 growing at 0.5 MPa
showed an increase of 58.7% in glutathione concentration in comparison with the control cells at
atmospheric pressure (Qiao et al., 2006a). Similarly, application of HP at 1 MPa on S. cerevisiae
resulted in increasing the yield of trehalose by 82.9 % (Dong et al., 2007). These two products
are usually present at low concentrations in microorganisms (Bachhawat et al., 2013), but their
5
production is stimulated under stressful conditions (such as HP), possibly as a mechanism for
protection (Dong and Jiang, 2016).
Table 1.1. General effects of high pressure on microbial cell growth and fermentation.
Microorganism Main effects References

- Acceleration of alcoholic
fermentation (up to 3-fold)
- Increased ethanol yields (5-6 %)
- Over expression of several genes Bravim et al. (Bravim et al.,
related to cell recovery and stress 2012); Dong et al. (2007);
Saccharomyces cerevisiae
tolerance, including the gene Picard et al. (2007); Qiao et al.
SYM1 (2006b)
- Increased glutathione
concentration (58.7 %)
- Increased trehalose yield (82.9 %)
- Metabolic shift, with increased
ethanol production and decreased
Clostridium thermocellum Bothun et al. (2004)
acetate (by-product) production
- Higher ethanol:acetate ratio
- Decreased cell growth
- Decreased production of bacterial
Gluconacetobacter xylinus cellulose Kato et al. (2007)
- Cellulose ribbons with profound
morphological differences
- Decreased fermentation rate
- At higher pressures,
Streptococcus thermophilus, microorganisms metabolically
Lactobacillus bulgaricus, and inhibited Mota et al. (2015)
Bifidobacterium lactis - Bacterial strains still viable at
lower pressures, with ability to
produce yogurt
- Fermentation with O. oeni during
and after HP-stresses
- Decreased concentrations of L-
Oenococcus oeni Neto et al. (2016)
lactate
- Increased concentrations of D-
lactate
Thermophilic bacterium Clostridium thermocellum efficiently converts cellobiose to

biofuels and other chemicals, but also synthesizes several undesired co-products, such as organic
acids (acetate, lactate) and gaseous end-products (H2, CO2), limiting the commercialization of
biofuel produced by this microorganism (Béguin and Aubert, 1994; Bothun et al., 2004; Herrero
et al., 1985; Wiegel, 1980). Application of pressure of 7 and 17 MPa on C. thermocellum
redirected the fermentative metabolism from the production of organic acids (such as acetate) to
ethanol, leading to a 60-fold increase in the ratio ethanol:acetate under HP compared to
6
atmospheric pressure. The authors explained these shifts with the increased concentration of
dissolved gases at HP and the consequent modification of the metabolic pathways involving these
product gases (Bothun et al., 2004).
High pressure might also be used to alter the properties of biopolymers produced by
microorganisms (Aertsen et al., 2009). Gluconacetobacter xylinus has been used as a model
organism for bacterial cellulose biosynthesis (Ross et al., 1991). It was demonstrated that G.
xylinus cells remained viable and retained the ability to produce cellulose under HP (30, 60 and
100 MPa). However, a decrease in the count of viable cells was observed at 100 MPa, leading to
a decrease in the amount of cellulose produced. Cellulose ribbons had profound morphological
differences depending on the applied pressure conditions, since the cellulose fibers produced
under HP had significantly higher density compared with the cellulose produced at atmospheric
pressure (Kato et al., 2007). Nevertheless, the result of these morphological changes on the
functional properties of the polymer still needs further investigation.
Recently, our research group evaluated the effect of HP (5-100 MPa) on lactate
fermentation that occurs during probiotic yogurt production (Mota et al., 2015). We observed that
HP reduced the fermentation rate, possibly due to the inhibitory effect of pressure on the growth
of the microorganisms relevant to the process (Streptococcus thermophilus, Lactobacillus
bulgaricus, and Bifidobacterium lactis). However, extension of the fermentation time at 5 MPa
(used as a case study) yielded a typical pH for probiotic yogurt, showing that lactic acid
fermentation can be performed at this pressure without compromising the viability of the bacterial
strains (including the probiotic one) and their ability to produce yogurt. However, more studies
are needed to understand the effects of HP on the nutritional and organoleptic properties of yogurt
produced at these conditions.
It is also possible to apply HP-stresses only in the beginning of fermentation, with the
remaining time taking place at atmospheric pressure. Neto et al. (2016) used this approach when
evaluating the effect of different HP-stresses (50 MPa, 8 h; 100 MPa, 8 h; 300 MPa, 0.5 h) on
Oenococcus oeni metabolism. This is a lactic acid bacterium employed by wine industry to
perform malolactic fermentation. Oenococcus oeni was able to perform fermentation during and
after HP-stresses of 50 and 100 MPa, although with some metabolic changes. Particularly, the
HP-stress of 100 MPa lead to lower concentrations of L-lactate and higher concentrations of D-
lactate, compared to the control. The HP-stress of 300 MPa, 0.5 h resulted in complete inactivation
of O. oeni, but malolactic fermentation was still observed at some extent, showing that malolactic
enzyme maintained some residual activity at these conditions.
All in all, the studies in literature reveal that HP could not only be applied intermittently,
but can also be maintained during the whole fermentation time, without serious cell loss and no
heating effect. Since there is no need of refrigeration, the energetic costs of the fermentative
7
process are lower, and the application of HP is simpler. This versatility can represent an advantage
to HP-assisted fermentation, since it allows to adapt the application mode according to the
specificities and requirements of each bioprocess.
1.2.1. Technological equipment suitable for fermentation processes

In this section, the most relevant aspects regarding HP-equipment for microbial growth
and fermentation are briefly described. Since the 1990s, a number of equipment manufacturers
have entered the HP industry, most of them starting from parallel sectors (such as water jet cutting
and diamond manufacturing) with their experience. HP equipment and services industry have
been expanded and have spread globally. At present, companies like HIPERBARIC and AVURE
Technologies are dominating the market of industrial HP systems (Elamin et al., 2015).
It is difficult to compare different types of HP equipment because each one offers unique
characteristics based on the operation parameters, such as range of operating pressure,
temperature control, volume of the vessel or layout of system (Bermúdez-Aguirre and Barbosa-
Cánovas, 2011). High pressure systems are typically made up of high strength steel alloys with
high fracture toughness and corrosion resistance. The most suitable equipment for fermentation
(Figure 1.1) includes a fermenter in a pressure vessel (thick-wall cylinder), in some cases with
agitation, aeration, and temperature control.
The desired pressure in the pressure vessel is achieved through compression of pressure-
transmitting fluid, using the combined action of a pump and an intensifier. The most commonly
used pressure-transmitting fluid is water, but glycol, glycol and water, silicone oil, sodium
benzoate solution or castor oil may also be used (Balasubramaniam et al., 2015). The specificities
of the system may be dependent on the type of fermentation and the requirements of the microbial
strain. For instance, aeration and agitation are highly beneficial for aerobic processes, while it can
be dispensable for anaerobic fermentations. On the other hand, temperature control is usually an
essential feature of the equipment, since even slight variation in temperature may be enough to
compromise microbial cell growth and fermentation. The pressure range allowed by this type of
pressure systems is usually low, since these fermentation applications do not require high pressure
levels.
It is important to consider that it may be unpractical (and expensive) to acquire a HP
equipment that is specific for cell growth and fermentation experiments. Therefore, some of the
studies in literature were performed in different equipment layouts, with general features adapted
for a wide range of HP experiments (food pasteurization and preservation, extractions, and many
others). In order to perform fermentation studies in this type of systems, extensive process
8
optimization must be carried out, in attempt minimize the hurdles related to lack of agitation,
aeration or temperature control.
Figure 1.1. Schematic representation of microbial fermentation under high pressure used at laboratory
scale. Reproduced from Mota et al. (2018) with permission of Taylor & Francis Group LLC in the format
Thesis/Dissertation via Copyright Clearance Center.
High pressure systems have been highly studied and engineered in the last decades,
resulting in considerable advances in the field. However, some challenges still remain, as the need
to improve the strength of HP vessels and the capacity of the pumps, as well as the vessel’s
resistance to a high number of cycles. On the other hand, the main issue related to HP equipment
for cell growth/fermentation is the adaptation to the particular and unique requirements of each
microorganism and process. Therefore, these HP systems need to be highly adjustable, through
the utilization of different modules and functions that should be adapted for each fermentation
process.
1.3. Comparison between high pressure and other non-conventional technologies

Although many differences were observed between the technologies tested, all of them seem
to be suitable for application (under specific conditions) during the fermentative processes.
Electric fields are able to promote beneficial effects to fermentation, but a careful optimization of
process parameters is necessary to avoid loss of cell viability, since the application of electric
fields (particularly in the case of high intensity PEF) are usually very aggressive to the cells (Barba
9
et al., 2015a; Varghese et al., 2014). In addition, the scaling up of PEF systems, their high initial
cost, and the availability of commercial units are relevant drawbacks of this technology (Morales-
de La Peña et al., 2011). In contrast, US are a well-studied technology, with several applications
in different fields, and with lower equipment costs. A further benefit is that US is usually less
aggressive to the cells, making this technology more feasible for application on fermentation
processes (Chisti, 2003; Gogate and Kabadi, 2009). However, the application of electric fields or
ultrasonication during fermentation implies an increase of temperature that affects the viability of
the microbial strains and compromises the entire fermentative process. As a result, a refrigeration
system is needed to control the temperature during the process, leading to higher energy
requirements and thus higher energetic costs.
Therefore, HP seems to be particularly suitable for application to fermentation processes,
since it is possible to use it continuously without inactivation problems or heat generation, in
contrast to the other technologies. The application of sub-lethal HP has higher versatility, as it
can be applied as stresses or continuously, without heating effects, leading to a higher variety of
possibilities. Additionally, HP only needs energy to generate the pressure and more energy is
needed to maintain the pressure, and so application of HP stress during the whole fermentation
process has also the advantage of not involving energetic costs.
Moreover, HP has interesting effects on living systems, offering great biotechnological
potential. Such an example is the thriving of living organisms under deep-sea, since pressure
increases with depth in the oceans, at an approximate rate of 10 MPa per km in the water column.
The microorganisms living under these environments can be a valuable source of compounds with
interesting biological activity and useful enzymes with novel properties. Consequently, these
extremophiles are attracting the attention of several pharmaceutical companies, wishing to obtain
novel compounds with biological activity. In fact, marine natural products used to drug
development cover a very wide range of pharmacological effects (Newman and Cragg, 2004).
From this perspective, fermentation under HP can be used to mimic the conditions present in
deep-sea environments, challenging the microorganisms to respond and adapt, with possible
production of several added-value products. Despite of the strong interest of pressure application
on biological processes, the information in literature is very limited, possibly due to the low
availability of HP systems and the high cost of the equipment, which correspond to the main
challenges to a more widespread utilization of this approach.
Overall, the performance of fermentation under non-conventional conditions is still poorly
explored, but the studies published so far suggest potential development within the next years,
due to the wide biotechnological applications. At this point it is difficult to assess the economical
feasibility of these methods, since there are many factors affecting the processes that still need
better understanding and optimization. In addition, since most studies were performed at
10
laboratory-scale, the scale-up of these processes will be an important requirement to their

implementation. Therefore, we expect that further work conducted in this field will focus on
addressing these issues, as well as on the evaluation of these technologies on other fermentative
processes with industrial relevance, mainly for non-food applications, which are considerably
scarcer within this topic.
11
1.4. References
Abe, F., 2007. Exploration of the effects of high hydrostatic pressure on microbial growth,
physiology and survival: Perspectives from piezophysiology. Biosci. Biotechnol. Biochem.
71, 2347–2357.
Aertsen, A., Meersman, F., Hendrickx, M.E.G., Vogel, R.F., Michiels, C.W., 2009.
Biotechnology under high pressure: Applications and implications. Trends Biotechnol. 27,
434–441.
Bachhawat, A.K., Thakur, A., Kaur, J., Zulkifli, M., 2013. Glutathione transporters. Biochim.
Biophys. Acta - Gen. Subj. 1830, 3154–3164.
Barba, F.J., Esteve, M.J., Frígola, A., 2012. High pressure treatment effect on physicochemical
and nutritional properties of fluid foods during storage: A review. Compr. Rev. Food Sci.
Food Saf.
Barba, F.J., Parniakov, O., Pereira, S.A., Wiktor, A., Grimi, N., Boussetta, N., Saraiva, J.A., Raso,
J., Martin-Belloso, O., Witrowa-Rajchert, D., 2015a. Current applications and new
opportunities for the use of pulsed electric fields in food science and industry. Food Res.
Int. 77, 773–798.
Barba, F.J., Terefe, N.S., Buckow, R., Knorr, D., Orlien, V., 2015b. New opportunities and
perspectives of high pressure treatment to improve health and safety attributes of foods. A
review. Food Res. Int. 77, 725–742.
Bartlett, D.H., 2002. Pressure effects on in vivo microbial processes. Biochim. Biophys. Acta -
Protein Struct. Mol. Enzymol. 1595, 367–381.
Béguin, P., Aubert, J.P., 1994. The biological degradation of cellulose. FEMS Microbiol. Rev.
13, 25–58.
Bermúdez-Aguirre, D., Barbosa-Cánovas, G. V, 2011. An update on high hydrostatic pressure,
from the laboratory to industrial applications. Food Eng. Rev. 3, 44–61.
Bothun, G.D., Knutson, B.L., Berberich, J.A., Strobel, H.J., Nokes, S.E., 2004. Metabolic
selectivity and growth of Clostridium thermocellum in continuous culture under elevated
hydrostatic pressure. Appl. Microbiol. Biotechnol. 65, 149–157.
Bravim, F., Lippman, S.I., da Silva, L.F., Souza, D.T., Fernandes, A.A.R., Masuda, C.A., Broach,
J.R., Fernandes, P.M.B., 2012. High hydrostatic pressure activates gene expression that
leads to ethanol production enhancement in a Saccharomyces cerevisiae distillery strain.
Appl. Microbiol. Biotechnol. 97, 2093–2107.
Chisti, Y., 2003. Sonobioreactors: using ultrasound for enhanced microbial productivity. Trends
Biotechnol. 21, 89–93.
De Angelis, M., Bini, L., Pallini, V., Cocconcelli, P.S., Gobbetti, M., 2001. The acid-stress
response in Lactobacillus sanfranciscensis CB1. Microbiology 147, 1863–1873.
12
Dong, Y., Jiang, H., 2016. Microbial production of metabolites and associated enzymatic
reactions under high pressure. World J. Microbiol. Biotechnol. 32, 178.
Dong, Y., Yang, Q., Jia, S., Qiao, C., 2007. Effects of high pressure on the accumulation of
trehalose and glutathione in the Saccharomyces cerevisiae cells. Biochem. Eng. J. 37, 226–
230.
Elamin, W.M., Endan, J.B., Yosuf, Y.A., Shamsudin, R., Ahmedov, A., 2015. High pressure
processing technology and equipment evolution: A review. J. Eng. Sci. Technol. Rev. 8.
Galván-D’Alessandro, L., Carciochi, R.A., 2018. Fermentation assisted by pulsed electric field
and ultrasound: A review. Fermentation 4, 1.
Gogate, P.R., Kabadi, A.M., 2009. A review of applications of cavitation in biochemical
engineering/biotechnology. Biochem. Eng. J. 44, 60–72.
Herrero, A.A., Gomez, R.F., Roberts, M.F., 1985. 31P NMR studies of Clostridium
thermocellum. Mechanism of end product inhibition by ethanol. J. Biol. Chem. 260, 7442–
7451.
Hill, C., Cotter, P.D., Sleator, R.D., Gahan, C.G.M., 2002. Bacterial stress response in Listeria
monocytogenes: jumping the hurdles imposed by minimal processing. Int. Dairy J. 12, 273–
283.
Hörmann, S., Scheyhing, C., Behr, J., Pavlovic, M., Ehrmann, M., Vogel, R.F., 2006.
Comparative proteome approach to characterize the high-pressure stress response of
Lactobacillus sanfranciscensis DSM 20451T. Proteomics 6, 1878–1885.
Huang, H.-W., Lung, H.-M., Yang, B.B., Wang, C.-Y., 2014. Responses of microorganisms to
high hydrostatic pressure processing. Food Control 40, 250–259.
Kato, N., Sato, T., Kato, C., Yajima, M., Sugiyama, J., Kanda, T., Mizuno, M., Nozaki, K.,
Yamanaka, S., Amano, Y., 2007. Viability and cellulose synthesizing ability of
Gluconacetobacter xylinus cells under high-hydrostatic pressure. Extremophiles 11, 693–
698.
Lado, B.H., Yousef, A.E., 2002. Alternative food-preservation technologies: Efficacy and
mechanisms. Microbes Infect. 4, 433–440.
Lou, Y., Yousef, A.E., 1997. Adaptation to sublethal environmental stresses protects Listeria
monocytogenes against lethal preservation factors. Appl. Environ. Microbiol. 63, 1252–
1255.
Macgregor Jr, R.B., 2002. The interactions of nucleic acids at elevated hydrostatic pressure.
Biochim. Biophys. Acta - Protein Struct. Mol. Enzymol. 1595, 266–276.
Mills, S., Stanton, C., Fitzgerald, G.F., Ross, R.P., 2011. Enhancing the stress responses of
probiotics for a lifestyle from gut to product and back again. Microb. Cell Fact. 10, S19.
Morales-de La Peña, M., Elez-Martínez, P., Martín-Belloso, O., 2011. Food preservation by
13
pulsed electric fields: An engineering perspective. Food Eng. Rev. 3, 94–107.

Mota, M.J., Lopes, R.P., Delgadillo, I., Saraiva, J.A., 2015. Probiotic yogurt production under
high pressure and the possible use of pressure as an on/off switch to stop/start fermentation.
Process Biochem. 50, 906–911.
Mota, M.J., Lopes, R.P., Delgadillo, I., Saraiva, J.A., 2013. Microorganisms under high pressure
- Adaptation, growth and biotechnological potential. Biotechnol. Adv. 31, 1426–1434.
Mota, M.J., Lopes, R.P., Koubaa, M., Roohinejad, S., Barba, F.J., Delgadillo, I., Saraiva, J.A.,
2018. Fermentation at non-conventional conditions in food-and bio-sciences by the
application of advanced processing technologies. Crit. Rev. Biotechnol. 38, 122–140.
Neto, R., Mota, M.J., Lopes, R.P., Delgadillo, I., Saraiva, J.A., 2016. Growth and metabolism of
Oenococcus oeni for malolactic fermentation under pressure. Lett. Appl. Microbiol. 63,
426–433.
Newman, D.J., Cragg, G.M., 2004. Marine natural products and related compounds in clinical
and advanced preclinical trials. J. Nat. Prod. 67, 1216–1238.
Niven, G.W., Miles, C.A., Mackey, B.M., 1999. The effects of hydrostatic pressure on ribosome
conformation in Escherichia coli: An in vivo study using Differential Scanning Calorimetry.
Microbiology-(UK) 145, 419–425.
Patterson, M.F., 2005. Microbiology of pressure-treated foods. J. Appl. Microbiol. 98, 1400–
1409.
Picard, A., Daniel, I., Montagnac, G., Oger, P., 2007. In situ monitoring by quantitative Raman
spectroscopy of alcoholic fermentation by Saccharomyces cerevisiae under high pressure.
Extremophiles 11, 445–452.
Qiao, C., Liu, B., Xu, X., Jia, S., 2006a. The changes of glutathione and ergosterol in
Saccharomyces cerevisiae under high pressure. J. Chinese Biotechnol. 26, 59.
Qiao, C., Liu, B., Xu, X., Jia, S., 2006b. The changes of glutathione and ergosterol in
Saccharomyces cerevisiae under high pressure. J. Chinese Biotechnol. 26, 56–59.
Ross, P., Mayer, R., Benziman, M., 1991. Cellulose biosynthesis and function in bacteria.
Microbiol. Mol. Biol. Rev. 55, 35–58.
Serrazanetti, D.I., Guerzoni, M.E., Corsetti, A., Vogel, R., 2009. Metabolic impact and potential
exploitation of the stress reactions in lactobacilli. Food Microbiol. 26, 700–711.
Storz, G., Hengge, R., 2000. Bacterial Stress Responses. American Society for Microbiology
Press, Washington DC, US.
van de Guchte, M., Serror, P., Chervaux, C., Smokvina, T., Ehrlich, S.D., Maguin, E., 2002. Stress
responses in lactic acid bacteria. Antonie Van Leeuwenhoek 82, 187–216.
Varghese, K.S., Pandey, M.C., Radhakrishna, K., Bawa, A.S., 2014. Technology, applications
and modelling of ohmic heating: A review. J. Food Sci. Technol. 51, 2304–2317.
14
Welch, T.J., Farewell, A., Neidhardt, F.C., Bartlett, D.H., 1993. Stress response of Escherichia
coli to elevated hydrostatic pressure. J. Bacteriol. 175, 7170–7177.
Wemekamp-Kamphuis, H.H., Wouters, J.A., de Leeuw, P.P.L.A., Hain, T., Chakraborty, T.,
Abee, T., 2004. Identification of sigma factor σB-controlled genes and their impact on acid
stress, high hydrostatic pressure, and freeze survival in Listeria monocytogenes EGD-e.
Appl. Environ. Microbiol. 70, 3457–3466.
Wiegel, J., 1980. Formation of ethanol by bacteria. A pledge for the use of extreme thermophilic
anaerobic bacteria in industrial ethanol fermentation processes. Experientia 36, 1434–1446.
Winter, R., Jeworrek, C., 2009. Effect of pressure on membranes. Soft Matter 5, 3157–3173.
Zamfir, M., Grosu-Tudor, S.-S., 2014. Stress response of some lactic acid bacteria isolated from
Romanian artisan dairy products. World J. Microbiol. Biotechnol. 30, 375–384.
15
16
CHAPTER II
Fermentation of glycerol for production of

valuable bio-chemicals: A literature revision
CHAPTER II. Fermentation of glycerol for production of valuable bio-chemicals
2.1. The biodiesel industry and the glycerol surplus

Global energy demand is met by petrochemical sources, coal and natural gases. These
sources are finite and the current high consumption rate may lead to depletion of fossil fuels. In
addition to that, it is necessary to make great efforts to reduce emissions of sulphur, nitrogen and
carbon oxides, which are released from the fossil fuel combustion and correspond to a
fundamental source of global warming (Vivek et al., 2017). For these ecological and economic
reasons, there is a growing awareness to look for fuels derived from renewable sources, such as
biomass. In that context, biodiesel is one the most promising alternatives to fossil fuels, since it
is renewable, biodegradable, environmentally-friendly, and can be used directly in diesel engines
without major modifications (da Silva et al., 2009; McNutt et al., 2017). It is manufactured from
vegetable, plant derived oils or animal fat as feedstock, consisting of long chain alkyl esters.
Biodiesel is chemically synthesized by transesterification of lipids with an alcohol in the presence
of catalyst resulting in a mono alkyl ester (Vivek et al., 2017).
In United States and Europe, biodiesel is a fast growing product as their governments
policies seek to spur the development of renewable transportation fuels (Lee et al., 2015). Global
biodiesel production has significantly increased annually, achieving 22.7 million tons in 2012,
and it is forecasted to reach 36.9 million tons by 2020 (Ciriminna et al., 2014; Katryniok et al.,
2013; OECD, 2011; Sun et al., 2017). The fast growth of the biodiesel industry has created
environmental and sustainability issues, such as the formation of glycerol as the major reaction
by-product: in the transesterification process, 3 mol of fatty acid methyl esters and 1 mol of
glycerol are synthesized. The glycerol by-product yields approximately 10 wt% of the total
product (Lee et al., 2015). Due to the current increasing biofuel demand, worldwide production
of glycerol climbed from 1 million tons in 2000 to 3 million tons in 2011 and is expected to reach
6 million tons by 2025 (Ciriminna et al., 2014; Katryniok et al., 2013; OECD, 2011; Sun et al.,
2017). The increased availability of glycerol has caused a substantial reduction in its cost.
According to a recent report, the price of crude glycerol is $0.24/kg and that of pure grade glycerol
is $0.90/kg (Kumar and Park, 2018; San Kong et al., 2016). Effective and economic utilization of
the enormous amounts of glycerol produced by the biodiesel industry is still very challenging
(Khanna et al., 2012). Glycerol is traditionally used for many applications in the cosmetic, paint,
automotive, food, tobacco, pharmaceutical, pulp and paper, leather and textile industries (da Silva
et al., 2009). However, it is essential to find alternative and sustainable applications for glycerol.
Extensive research has been conducted to investigate ways to utilize this surplus crude
glycerol. The annual number of research articles addressing the use of glycerol has increased to
>7000, doubling in number from the year 2000 to 2007 (Kumar and Park, 2018). One of the most
popular approaches is the conversion of glycerol into high-value products by either biological or
chemical transformations (Mattam et al., 2013). However, biological conversions are generally
19
preferred to chemical processes, due to the higher specificity of the reaction, lower temperature
and pressure involved in the reaction process, and lower levels of chemical contaminants (Johnson
and Taconi, 2007; Mattam et al., 2013; Vivek et al., 2017). The use of glycerol as substrate for
microbial fermentation processes will be addressed in detail in the following sections.
2.2. Glycerol bio-conversion into valuable products

In the last decade, significant research efforts have been focusing on the use of glycerol
as substrate for the production of a wide variety of valuable bio-chemicals, such as organic acids,
alcohols, lipids and polymers (Mattam et al., 2013; Vivek et al., 2017). Some of the main products
obtained from glycerol fermentation are indicated in Table 2.1, as well as the microbial strains
usually involved in those processes.
Table 2.1. Some of the bioproducts obtained from glycerol in fermentation by different microorganisms.
Bioproduct Main producing-microorganisms References
Actinobacillus succinogenes;
Carvalho et al. (2014); Lee et al.
Succinate Anaerobiospirillum succiniciproducens;
(2001); Scholten et al. (2009)
Basfia succiniciproducens
Citrate Yarrowia lipolytica Rywińska and Rymowicz (2010)
Propionibacterium acidipropionici;
Propionibacterium acnes; Clostridium Barbirato et al. (1997); Wang et al.
Propionate
propionicum; Propionibacterium (2015); Zhu et al. (2010)
freudenreichii subsp. Shermanii
Escherichia coli; Lactobacillus rhamnosus; Hong et al. (2009); Murakami et al.
Lactate
Enterobacter faecalis (2016); Prada-Palomo et al. (2012)
Clostridium butyricum; Klebsiella Boenigk et al. (1993); Grahame et
pneumonia; Citrobacter freundii; al. (2013); Lüthi-Peng et al. (2002);
1,3-Propanediol Lactobacillus reuteri; Lactobacillus brevis; Menzel et al. (1997); Petitdemange
Lactobacillus diolivorans; Lactobacillus et al. (1995); Pflügl et al. (2012);
panis Vivek et al. (2016)
Klebsiella pneumonia; Klebsiella oxytoca; Biebl et al. (1998); Metsoviti et al.
2,3-Butanediol
Bacillus amyloliquefaciens (2012); Yang et al. (2013)
Butanol Clostridium pasteurianum Biebl (2001)
Ashby et al. (2004); Bormann and
Cupriavidus necator; Burkholderia sacchari;
Roth (1999); Huijberts et al.
a Pseudomonas oleovorans; Pseudomonas
PHA (1992); Ibrahim and Steinbüchel
putida; Zobellella denitrificans; Paracoccus
(2010); Mothes et al. (2007);
denitrificans
Rodríguez-Contreras et al. (2015)
a
PHA: Polyhydroxyalkanoates
20
Because of its reduced nature, glycerol generates twice the number of reducing
equivalents per carbon than do traditional carbohydrates (glucose, xylose, and sucrose) when
converted into the glycolytic intermediates phosphoenolpyruvate (PEP) or pyruvate (Dharmadi et
al., 2006). As a consequence, glycerol gives higher yields of reduced metabolites (e.g. succinate,
ethanol, 1,3-propanediol, and butanol) than does glucose (Kumar and Park, 2018; Yazdani and
Gonzalez, 2007) (Table 2.2).
Table 2.2. Comparison of maximum theoretical yield for production of different compounds from
glycerol and glucose. Adapted from Kumar and Park (2018).
a
Maximum theoretical yield
Bioproduct
Glucose (mol/0.5 mol) Glycerol (mol/mol)
1,3-Propanediol 0.75 0.88
2,3-Butanediol 0.55 0.64
Ethanol 1.00 1.17
Succinate 0.86 1.00
Lactate 1.00 1.17
Acetate 1.50 1.75
Butanol 0.50 0.58

aThe theoretical yields of metabolites were calculated on the basis of degree of reduction (Dugar and Stephanopoulos,
2011)
Although glycerol conversion is highly variable according to the microorganism, it is

possible to define generalized metabolic pathways that are usually present in most glycerol-
consumer strains. The initial step corresponds to the uptake of glycerol into the cells, which can
happen passively since glycerol is a small and uncharged molecule, and thus permeable across
the membranes (Johnson et al., 2016). However, cells that are limited to passive uptake have a
growth disadvantage at low glycerol concentrations (da Silva et al., 2009). Facilitated diffusion
can be achieved by an integral membrane protein, the glycerol facilitator GlpF (Darbon et al.,
1999; Heller et al., 1980; Voegele et al., 1993), which is able to selectively diffuse glycerol across
the cellular membranes (Fu et al., 2000).
The generalized metabolism of glycerol in microorganisms was reviewed by Khanna et
al. (2012). It mainly occurs by two distinct and parallel pathways, oxidative and reductive
(Yazdani and Gonzalez, 2007), both shown in Figure 2.1. Some microorganisms, such as the
members of Enterobacteriaceae family, such as Klebsiella, Citrobacter and Clostridium exhibit
both oxidative and reductive pathways.
21
Figure 2.1. Metabolic pathway of glycerol bioconversion: oxidative and reductive wings of glycerol
assimilation (E1–Glycerol dehydratase; E2–1,3 PDO dehydrogenase; E3–Aldehyde dehydrogenase; E4–
Glycerol dehydrogenase type 1; E5–Dihydroxy acetone kinase; E6–Triose phosphate isomerase; E7–
Lactate dehydrogenase; E8–Pyruvate decarboxylase; E26–Pyruvate kinase; E27–α-Acetolactate
decarboxylase; E28–Acetoin reductase; E29–Diacetyl reductase; E31–Glycerol 3-phosphate
dehydrogenase). Reproduced from Khanna et al. (2012) with permission of Taylor & Francis Group LLC
in the format Thesis/Dissertation via Copyright Clearance Center.
The reductive pathway proceeds by the dehydration of glycerol to 3-

hydroxypropionaldehyde (3-HPA), also called as reuterin, in presence of glycerol dehydratase.
The 3-HPA is then reduced to 1,3-propanediol by a NADH-dependant 1,3-propanediol
dehydrogenase, regenerating NAD+ for the oxidative branch. 3-Hydroxypropionaldehyde can be
further oxidized to 3-hydroxypropionic acid by NAD-linked aldehyde dehydrogenase (Khanna et
al., 2012).
A parallel oxidative branch produces dihydroxyacetone and ultimately channels glycerol
to Embden–Meyerhof–Parnas (EMP) pathway for the production of a number of important
metabolites, such as ethanol, citrate, succinate, lactate, 2,3-butanediol, and others. The oxidative
wing progresses by the oxidation of glycerol to dihydroxyacetone (DHA) by a NAD-linked type
1 glycerol dehydrogenase, that is further phosphorylated by an ATP-dependant DHA kinase to a
glycolytic intermediate, DHA phosphate, which enters EMP pathway. The production of DHA
22
from glycerol oxidation is primarily observed in acetic acid bacteria, Acetobacter and
Gluconobacter. These microorganisms are anaerobes, and hence, they lack glycolytic and Kreb’s
cycle enzymes and, as a consequence, further metabolism of DHA phosphate is not carried out
by them (Khanna et al., 2012).
In aerobe microorganisms, DHA phosphate isomerizes to glyceralehyde-3-phosphate by
triose phosphate isomerase. Glyceraldehyde-3-phosphate converts to phosphoenolpyruvate (PEP)
through a series of steps and finally undergoes substrate level phosphorylation to form pyruvate
by pyruvate kinase, generating ATP in the process. The pyruvate formed in the earlier step can
follow different metabolic pathways (Khanna et al., 2012):
i) It can be reduced to lactate by NADH-dependant lactate dehydrogenase;
ii) It may undergo decarboxylation by pyruvate decarboxylase to form α-acetolactate,
which is further decarboxylated to acetoin. α-Acetolactate may also decarboxylate
to diacetyl, which in turn is reduced to acetoin by NADPH-dependant diacetyl
reductase. The acetoin so formed, through any of the two routes, is reversibly
reduced to 2,3-butanediol by NADH-dependant acetoin reductase (Syu, 2001);
iii) It may convert into formate in presence of pyruvate formate lyase, which is further
cleaved to gaseous products H2 and CO2 by formate hydrogen lyase;
iv) It can be carboxylated by pyruvate carboxylase to oxaloacetate, which is further
reduced by NADH-dependant malate dehydrogenase to malate. Malate then
gradually forms fumarate by fumarase, and fumarate converts into succinate by
fumarate reductase;
v) It can form acetyl CoA in presence of pyruvate dehydrogenase complex. Acetyl
CoA is the precursor for the formation of acetate, butyrate, butanol and ethanol.
Acetyl CoA can also enter in lipid biosynthesis pathway to produce triacylglycerols,
also known as single cell oil.
Although the complete assimilation pathway of glycerol is rather complex, every step
yields a particular product. The range of metabolites varies according to the microorganism and,
particularly, the presence of an enzyme or a group of enzymes (Khanna et al., 2012).
2.3. Glycerol bio-conversion into 1,3-propanediol

1,3-Propanediol (1,3-PDO) is a diol monomer well known for numerous applications in
cosmetics, solvents, adhesives, detergents, and resins (Vivek et al., 2017). Recently, this monomer
has gained much attention in production of biodegradable polyester polytrimethylene
terephthalate (PTT), having significant application in carpet and textile industry (Kaur et al.,
23
2012; Traub et al., 1995). It is also a GRAS (Generally Recognized as Safe) food and beverage
ingredient, commercialized by DuPont Tate & Lyle Bio Products. It can be used as a flavor carrier,
to reduce bitterness and enhance sweetness, or as a humectant, or processing aid. It is approved
for use in different food products, such as seasonings and flavorings, confections and frostings,
nuts and nut products, alcoholic beverages, and frozen dairy products (DuPont Tate & Lyle Bio
Products, 2018). The global market of 1,3-PDO has increased considerably in recent years, due
to the increasing market demand of its derivatives. While the global demand for 1,3-PDO was
60.2 kilotons in 2012, it is expected to reach approximately 150 kilotons in 2019 (Lee et al.,
2015; Research TM, 2012).
The chemical synthesis of 1,3-PDO can be conducted by two main processes. The first
one is “Degussa” (hold by DuPont Company) and implies catalytically oxidation of propylene to
acrolein, which is hydrated next to 3-hydroxypropionaldehyde at medium pressure and
temperature, followed by the hydrogenation to 1,3-PDO using a rubidium catalyst at high
pressure. The second process carried out by Shell is based on oxidation of ethylene to ethylene
oxide, followed by production of 3-HPA through the reaction called “hydroformylation” (also
named “oxo synthesis”) at high pressures (around 150 bar). The aldehyde extraction from the
organic phase is performed using water, and the 3-HPA hydrogenation is conducted by using
nickel as a catalyst under high pressure (Mitrea et al., 2017). However, these processes have
numerous drawbacks, such as toxic intermediate products, the need to apply high pressures and
temperatures, expensive non-renewable petroleum based catalysts, as well as moderate process
efficiency and environmental noxiousness (Lin et al., 2005; Przystałowska et al., 2015; Raynaud
et al., 2003). Therefore, biotechnology offers the potential to develop and/or to improve the
production of these valuable compounds.
Currently, bio-based 1,3-PDO is produced by DuPont and Genencor International, Inc.,
by using a recombinant Escherichia coli strain, from glucose derived from corn as the sole carbon
source (Jolly et al., 2014; Maervoet et al., 2010). The biological conversion of glycerol into 1,3-
PDO is mainly attained by anaerobic bacteria fermentation process or micro-aerobic
fermentation (Chen et al., 2003; Liu et al., 2007; Yang et al., 2007). The bacterial fermentation of
glycerol to 1,3-PDO was first reported in 1881 by Zeng and Biebl (2002). Several members of
the genus Klebsiella (Ji et al., 2009), Citrobacter (Boenigk et al., 1993), Clostridium
(Petitdemange et al., 1995), Lactobacillus (Ricci et al., 2015), Enterobacter and Shimwellia
blattae (Rodriguez et al., 2016) are reported in literature as wild type producers of 1,3-PDO from
glycerol. Among all these microorganisms, Clostridium butyricum and Klebsiella pneumoniae
are usually considered the best natural producers. For instance, in anaerobic fermentation with
glycerol as the sole carbon source, Clostridium butyricum strains were observed to produce 1,3-
PDO titers higher than 60 g L-1 (Wilkens et al., 2012), and K. pneumoniae strains were reported
24
to produce titers between 60 and 90 g L-1 (Durgapal et al., 2014; Zhao et al., 2009). However,
Klebsiella family strains are classified as opportunistic pathogens and Clostridium family
members are obligate anaerobes. In consequence, strict process control is required in
fermentations with these microorganisms, thus making the process unattractive. Taking this into
account, the members of the Lactobacillus genus offer great potential for production of 1,3-PDO
from glycerol (Jolly et al., 2014). Some examples are indicated in Table 2.1, and include the
utilization of Lactobacillus reuteri, Lactobacillus panis, Lactobacillus diolivorans, and
Lactobacillus brevis for this purpose, achieving 1,3-PDO titers as high as 92 g L-1 (Lindlbauer et
al., 2017).
2.3.1. Production of 1,3-propanediol by Lactobacillus reuteri

Lactic acid bacteria (LAB) are a well-known group of microorganisms that have been
associated with humans throughout history. One important landmark was the first use of LAB as
starter cultures for dairy production in 1890, which marked the beginning of industrial
microbiology and established the industrial relevance of these microbial strains (Sauer et al.,
2017). There are numerous advantages in using fermentation by lactic acid bacteria (LAB) for
production of interest bio-based compounds. These include high carbon uptake rates, low biomass
formation, high stress tolerance, as well as strictly regulated simple metabolic pathways that lead
to a limited number of metabolites (Sauer et al., 2017). In addition, most members of LAB are
non-pathogenic and generally regarded as safe (GRAS) and, in consequence, these
microorganisms are not usually subjected to special legislative restrictions (Ricci et al., 2015).
Nevertheless, the utilization of LAB has remained mostly restricted to food and health,
while their potential for the bio-based chemical industry is still unexploited. Several studies report
that LAB have great potential for production of industrially valuable compounds, such as lactate,
mannitol, or 1,3-PDO (Sauer et al., 2017). In this context, Lactobacillus reuteri presents itself as
a LAB with high potential for bio-chemical production. It is a Gram-positive rod, facultative
anaerobe and obligate heterofermentative microbe (Lüthi-Peng et al., 2002). It is able to produce
1,3-PDO from glycerol, using the 5 enzymes encoded in its propanediol-utilization operon (pdu),
namely glycerol/diol dehydratase, propionaldehyde dehydrogenase, phosphotransacylase,
propionate kinase and 1,3-PDO oxidoreductase (Sriramulu et al., 2008).
Lactobacillus reuteri uses glycerol only as an electron acceptor and not as a carbon source
for growth. Even though L. reuteri is unable to grow on glycerol as the sole carbon source, the
strain can be used in a co-fermentation of glucose and glycerol (Ricci et al., 2015). The general
metabolic pathways of L. reuteri in a glycerol/glucose co-fermentation are represented in Figure
2.2.
25
Figure 2.2. Lactobacillus reuteri metabolic pathways from glucose and glycerol. Reprinted from Vieira et
al. (2015), with permission from Elsevier.
Glycerol enters the L. reuteri cell by the glycerol facilitator GlpF (Vivek et al., 2016) and
fermentation occurs in two steps. In the first step, glycerol is dehydrated into 3-HPA by a
coenzyme B12-dependent glycerol dehydratase (Burgé et al., 2015; Sardari et al., 2013a, 2013b;
Sriramulu et al., 2008; Talarico and Dobrogosz, 1990). The second step involves the reduction of
3-HPA to 1,3-PDO by a NADH-oxidoreductase, requiring NADH (El-Ziney et al., 1998; Schütz
and Radler, 1984; Talarico and Dobrogosz, 1989). Then, 3-HPA can also be oxidized into 3-
hydroxypropinonic acid (3-HP) through a three-step reaction (Dishisha et al., 2014; Luo et al.,
2011; Sabet-Azad et al., 2013). Sobolov and Smiley (1960) reported that 3-HPA was converted
to equimolar quantities of 3-HP and 1,3-PDO by an aldehydic dismutation of 3-HPA, particularly
when glycerol fermentation proceeded in the absence of sugar (glucose). On the other hand, 1,3-
PDO production was elevated and 3-HP production was reduced in the presence of glucose
concentrations (Kumar et al., 2013).
Therefore, glycerol/sugar co-fermentations are applied to stimulate cell growth and to
avoid the formation of 3-HP instead of 1,3-PDO. During glycerol/glucose co-fermentation,
glucose enters the glycolysis pathway, leading to the formation of various by-products, such as
ethanol, lactate, acetate, and others (Doleyres et al., 2005). Although both substrates are
metabolized by alternative pathways, the production of 1,3-PDO from glycerol is dependent on
26
glucose availability: as long as NADH is generated from glucose oxidative pathway, glycerol is
used as the preferred electron acceptor leading to its enhanced consumption (Jolly et al., 2014).
In addition, pyruvate obtained from the glycolysis pathway will compete with 3-HPA for NADH-
oxidoreductase to form other by-products such as lactate (El-Ziney et al., 1998).
Some of the most relevant studies on 1,3-PDO production by L. reuteri co-fermentation
(glycerol/sugar) are summarized in Table 2.3. In an attempt to increase 1,3-PDO production by
L. reuteri, Ragout et al. (1996) first tested a co-fermentation process with a sugar (maltose) and
glycerol. The results showed that L. reuteri changed its fermentation pattern to produce 4.4 g L-1
of 1,3-PDO as the major end product. Similarly, El-Ziney et al. (1998) observed that
glucose/glycerol co-fermentation resulted in the production of 3-HPA and 1,3-PDO, at the
expense of ethanol and lactate. More recently, several authors have been studying the co-
fermentation of glucose/glycerol by L. reuteri. Baeza-Jiménez et al. (2011) obtained 1,3-PDO
concentrations of 16.81 g L-1 in shake-flasks (100 mM glucose/200 mM glycerol), and 28.69 g L-
1
in bioreactor studies (200 mM glucose/400 mM glycerol). Jolly et al. (2014) reached a 1,3-PDO
concentration of 37.4 g L-1 after a batch fermentation with initial glucose and glycerol
concentrations of 20 and 63.5 g L-1, respectively. The authors also found that 1,3-PDO final
concentrations and yields under unaerated conditions were close to those obtained under
anaerobic conditions. Therefore, unaerated fermentation can be used to effectively produce 1,3-
PDO in bioreactors, which is highly beneficial for large scale fermentation processes, since
maintaining anaerobic conditions involves considerable costs.
Table 2.3. Main results of 1,3-PDO production of by L. reuteri, using glycerol as co-substrate.
Titer (g L-1) Yield (g g-1) Productivity (g L-1 h-1) References

4.4 - - Ragout et al. (1996)
4.4 - - El-Ziney et al. (1998)
28.7 0.91 1.05 Baeza-Jiménez et al. (2011)
65.3 0.67 1.2 Jolly et al. (2014)
a a a
9.8 0.70 4.92 Vieira et al. (2015)
46.0 0.74 0.66 Ricci et al. (2015)
2.5 0.36 0.07 Zaushitsyna et al. (2017)
a
For bioreactor operated at continuous mode
Another important factor for this fermentation is the concentration of acetate during the
process. Acetate is normally regarded as a product of anaerobic fermentation and it has been
recognized as a microbial growth inhibitor. However, Iino et al. (2002) have stated that certain
27
acetate concentrations may have a positive influence on growth and fermentation of LAB. Jolly
et al. (2014) observed that an initial acetate concentration of 5 g L-1 increased 1,3-PDO
productivity, while concentrations above this value were found to be inhibitory. A possible
explanation is related to the ability of acetate to act as an indirect proton and electron acceptor,
leading to the activation of oxidative pathway at early stages of fermentation. However, at higher
concentrations, acetate decreases the cells cytoplasmic pH. Since the cells expend energy to
restore cytoplasmic pH, less energy is available for growth and bacterial metabolism (Heyndrickx
et al., 1991; Jolly et al., 2014). Vieira et al. (2015) tested different bioreactor operation modes
(batch, repeated batch and continuous) and each one was found to affect L. reuteri glycerol
fermentation. The highest 1,3-PDO productivity (4.92 g L-1) was achieved at continuous mode.
Ricci et al. (2015) tested the ability of different L. reuteri strains for the production of
1,3-PDO in a glucose/glycerol co-fermentation, and concluded that only the strain DSM 20016
produced a considerable 1,3-PDO concentration (10 g L-1). In addition, this strain has some
specific features, such as the production of 1,3-PDO closely linked to biomass accumulation.
After optimization of substrate concentrations (0.4 M glucose, 0.8 M glycerol, in modified MRS
without sodium acetate) and process variables (37 ºC, 200 rpm, and pH 5.5), the authors reached
a 1,3-PDO concentration of 41 g L-1, which is the highest reported so far for a batch fermentation
with this microbial strain. Zaushitsyna et al. (2017) studied the use of L. reuteri as whole-cell
biocatalyst for immobilization and conversion of glycerol into 3-HPA, 3-HP and 1,3-PDO. With
that strategy, a 1,3-PDO concentration of 2.5 g L-1 was achieved.
Overall, the production of 1,3-PDO by L. reuteri from glycerol corresponds to a
sustainable and strategic biotechnological approach, mainly due to three different aspects:
i) The use of glycerol as one of the substrates of the process;
ii) The use of a non-pathogenic and safe fermentative microorganism;
iii) The production of 1,3-PDO, a bio-chemical intermediate with broad applications,
including in the production of biodegradable polyester polytrimethylene
terephthalate (PTT).
On the contrary, the current main limitations of the process correspond to:
i) The use of glucose as a co-substrate;
ii) The need to carefully adjust several process variables to promote 1,3-PDO
production.
Although there is no apparent straightforward strategy to overcome the former limitation,

the latter can be more easily addressed by analyzing all variables and parameters that may affect
the process, as well as their interaction. This would be the first step for further optimization of
this promising biotechnological process.
28
2.4. Glycerol bio-conversion into polyhydroxyalkanoates

Polyhydroxyalkanoates (PHA) represent a complex class of naturally occurring bacterial
polyesters that are synthesized intracellularly as carbon and energy reserve materials (Ashby et
al., 2004). These reserve materials are synthesized when the microorganisms are grown under
nutrient-limited conditions (Ashby et al., 2005), since they play a pivotal role in the long-term
survival of bacteria under stressful conditions (Tan et al., 2014). Polyhydroxyalkanoates possess
properties similar to various synthetic thermoplastics, such as polypropylene (Lee, 1996).
However, PHA shows some unique properties, including biodegradability, biocompatibility,
water resistance, and oxygen impermeability, and thus they can be used in a variety of disposable
packaging goods (Tanadchangsaeng and Yu, 2012). They are also great candidates for high-
valued applications in biomedical devices, e.g. surgical implants, sutures, meshes, or tissue
engineering scaffolds (Brigham and Sinskey, 2012; Chen and Wu, 2005; Williams et al., 1999;
Zinn et al., 2001). The intense research and commercial interest in PHA is evident from the rapid
increment in PHA-related publications. For instance, Web of Science citation report (Thomson
Reuters, New York, NY, USA) revealed that in the last 20 years, PHA-related documents have
increased by almost 10-fold, while citations have increased by more than 500-fold, corresponding
to an average citation count of about 1100 citations per year (Tan et al., 2014).
Industrial PHA production still has a high cost and cannot be economically competitive
with conventional plastics (Brämer et al., 2001). The main factors that increase PHA production
costs were reviewed by Choi and Lee (1999) and included PHA productivity, content and yield,
the cost of the carbon substrate, and the recovery method applied. Therefore, the use of superior
bacterial strains, inexpensive carbon sources, low-cost media, as well as efficient fermentation
conditions and easier downstream processing could render PHA a commercially viable product
(da Cruz Pradella et al., 2010). One of the most well studied strategies is the substitution of pure
sugars (such as glucose or sucrose) by cheaper carbon sources as basis feedstock (Rodríguez-
Contreras et al., 2015). Therefore, using glycerol as substrate for PHA production would be a
good way to recycle a biodiesel by-product, and simultaneously reduce the high production cost
of PHA (Tanadchangsaeng and Yu, 2012).
Depending upon the number of carbon atoms in the monomers, PHA are classified into
two distinct groups: scl-PHA (short chain length PHA), consisting of 3-5 carbon atoms; and mcl-
PHA (medium chain length PHA), consisting of 6-14 carbon atoms (Możejko-Ciesielska and
Kiewisz, 2016). In both cases, the PHA biosynthetic pathways are intricately linked with the
bacterium’s central metabolic pathways including glycolysis, Krebs Cycle, β-oxidation, de novo
fatty acids synthesis, amino acid catabolism, Calvin Cycle, and serine pathway (Lu et al., 2009;
Madison and Huisman, 1999; Peplinski et al., 2010; Rothermich et al., 2000; Shimizu et al., 2013;
Yamane, 1993). Anabolism of scl-PHA involves two enzymes that commonly condensate two
29
acetyl-CoA followed by the reduction of acetoacetyl-CoA to provide the ultimate substrate (R)-
3-hydroxybutyryl-CoA for the PHA synthase (PhaC) (Suriyamongkol et al., 2007). In contrast,
mcl-PHA production is closely linked to fatty acids metabolic routes: de novo fatty acid
biosynthesis with non-related carbon sources, and β-oxidation which is the main metabolic
pathway for related substrates like fatty acids (Shahid et al., 2013). In the former, the PHA
composition depends directly on the carbon source, while in the latter there is no relationship
between the carbon source and the resulting PHA composition.
The structural composition of PHA polymers depends not only on the carbon compound
supplied as the growth substrate, but also on the microbial strain used for that purpose (Możejko-
Ciesielska and Kiewisz, 2016). The PHA bioaccumulation trait is widespread among the bacterial
and archaeal domains with PHA-producing microbes occurring in more than 70 bacterial and
archaeal genera (Lu et al., 2009; Poli et al., 2011). It is predominantly investigated in Cupriavidus
necator (formerly Ralstonia eutropha), which can store PHA up to 96 % of its cell dry mass
(CDM) under conditions of nitrogen or phosphate limitation and excess of carbon source
(Verlinden et al., 2007). Therefore, this microorganism is the most popular to produce PHA from
glycerol. Other species commonly used for this process include bacteria from the genus
Pseudomonas and Burkholderia (Koller and Marsalek, 2015). In the last years, other PHA-
producers have been identified and evaluated. Some examples include Zobellella denitrificans
(Ibrahim and Steinbüchel, 2010) and Paracoccus denitrificans (Kalaiyezhini and Ramachandran,
2015; Mothes et al., 2007), both able to achieve high PHA accumulation from glycerol (up to 87
% and 72 %, respectively, at optimized conditions). Paracoccus denitrificans is a microbial strain
with interesting metabolic features, and it seems particularly suitable for growth under stressful
conditions. For instance, P. denitrificans showed not only ability to survive, but also robust
cellular growth, under extreme gravity accelerations - corresponding to 403,627 times g (the
normal acceleration resulting from gravity at the Earth's surface) (Deguchi et al., 2011). The main
findings regarding the production of PHA by P. denitrificans are briefly described in the following
section.
2.4.1. Polyhydroxyalkanoate production by Paracoccus denitrificans

Formerly known as Micrococcus denitrificans, this bacterium was first isolated from soil
more than one century ago by Beijerinck and Minkman (1910). It was then renamed to
Paracoccus denitrificans in 1969. This is a Gram-negative facultative methylotrophic bacterium,
able to synthesize PHA from many carbon sources, such as methanol, ethanol, 1-butanol, and 1-
pentanol (Ueda et al., 1992; Yamane et al., 1996a, 1996b). It exhibits metabolic versatility, and it
was shown to grow aerobically and anaerobically, performing complete or partial denitrification.
Bacterial denitrification play an important role in determining the fate of reactive nitrogen in both
30
terrestrial and aquatic environments, especially when oxygen is limiting (Thomson et al., 2012).
Denitrifying bacteria can switch between respiring oxygen and nitrate, using a process in which
nitrate is reduced (via nitrite, nitric oxide and nitrous oxide) to di-nitrogen in a series of reactions
(Giannopoulos et al., 2017).
Paracoccus denitrificans was found to accumulate a PHA copolymer, poly(3-
hydroxybutyrate-co-3-hydroxyvalerate), during growth on n-pentanol (Yamane et al., 1996b).
This strain was also reported to accumulate poly(3-hydroxybutyrate-co-3-hydroxyvalerate) when
methanol and n-amyl alcohol were added together to a nitrogen-limited medium (Ueda et al.,
1992). There are only few studies regarding the production of PHA from glycerol by P.
denitrificans that are summarized in Table 2.4.
Table 2.4. Main results of PHA production by P. denitrificans, using glycerol as substrate.
Titer Yield Productivity Content

References
(g L-1) (g g-1) (g L-1h-1) (% of cell dry weight)
- 0.5 - 70 Mothes et al. (2007)
Kalaiyezhini and
10.7 0.3 0.15 72
Ramachandran (2015)
a
24.2 - - 39.3 Kumar et al. (2018)
a
Study performed using Paracoccus sp. LL1, and not P. denitrificans
Mothes et al. (2007) published the first study on this subject, and obtained a maximum
specific growth rate of 0.25 h-1, which is higher than the observed in the same study for C. necator
(0.13 h-1). The molecular weight of poly(3-hydroxybutyrate) produced with P. denitrificans or
C. necator from crude glycerol varied between 620,000 and 750,000 Da. These values are lower
than the observed for the polymer produced from acetate or fructose (970,000 and 1,160,000 Da,
respectively). However, these values are sufficiently high for processing by common techniques
used in the polymer industry. The impact of common crude glycerol contaminants (NaCl or
K2SO4) on P. denitrificans growth and poly(3-hydroxybutyrate) production was evaluated. These
contaminants accumulate in the course of fermentation and could have an effect on growth and
poly(3-hydroxybutyrate) synthesis. When using crude glycerol containing 5.5 % NaCl, a
reduction in the poly(3-hydroxybutyrate) content (48 %) was observed. The poly(3-
hydroxybutyrate) yield coefficient was also reduced, obviously due to osmoregulation.
Interestingly, contamination with K2SO4 showed a less pronounced effect, and thus it seems to be
more tolerable than contamination with NaCl.
31
Kalaiyezhini and Ramachandran (2015) have also studied poly(3-hydroxybutyrate)

production from glycerol by P. denitrificans. In that case, the kinetics of poly(3-hydroxybutyrate)
biosynthesis was evaluated in a batch bioreactor, testing different operational parameters, such as
nitrogen source, carbon to nitrogen ratio, pH, aeration, and initial glycerol concentration. From
all the nitrogen sources tested in the study (yeast extract, urea, ammonium sulfate, ammonium
chloride, and ammonium nitrate), ammonium sulfate and yeast extracts were the most suitable
ones in shake-flask experiments. However, yeast extracts have showed superior results (compared
to ammonium sulfate) in bioreactor studies. In the case of oxygen transfer rate and initial glycerol
concentration, the most suitable conditions for bacterial growth were not the same as for polymer
production. While higher oxygen transfer rates promoted microbial growth, moderate oxygen
transfer rates promoted poly(3-hydroxybutyrate) accumulation in the cells. Similarly, optimal
glycerol concentration for cell growth was 40 g L-1, while it was 20 g L-1 for poly(3-
hydroxybutyrate) accumulation. The authors also reported that high initial carbon:nitrogen (C:N)
ratio favored polymer accumulation and its productivity. At a C:N ratio of 21.4 (mol mol -1), a
poly(3-hydroxybutyrate) concentration of 10.7 g L-1 was achieved, corresponding to 72 % of
CDM. Overall, the more suitable conditions for P. denitrificans for PHA production may be
carefully selected, since cell growth and polymer accumulation may not be stimulated at the same
culture conditions.
Recently, Kumar et al. (2018) reported the conversion of glycerol by a culture of
Paracoccus sp. LL1, isolated from the Lonar Lake in India, with production of poly(3-
hydroxybutyrate-co-3-hydroxyvalerate). The formation of this co-polymer by glycerol unveils
interesting features of Paracoccus metabolism. However, it is relevant to note that the specific
metabolic pathways behind glycerol conversion (into PHA and other products) by Paracoccus
strains are still not extensively studied and understood. Likewise, the metabolic responses of P.
denitrificans to stress are poorly documented, despite the evidences of its ability to survive and
grow under extreme conditions. In fact, the use of stress to modulate the diverse and complex
metabolism of P. denitrificans would potentially result in interesting effects, for the different
purposes that P. denitrificans is usually applied (i.e. denitrification and PHA production).
2.5. Current challenges and possible improvements

Nowadays, it is widely accepted that several products usually obtained from petroleum
can be produced biotechnologically from glycerol using specific microbial strains (da Silva et al.,
2009). As a consequence, significant research efforts are focusing on this subject, with a wide
range of microorganisms being tested and manipulated for glycerol consumption and conversion.
Despite of the countless advantages of these glycerol-based bioprocesses, their application at
32
industrial scale still faces many challenges. According to Khanna et al. (2012), there are different
aspects that need to be taken into account regarding the feasibility of glycerol bioconversion: the
fluctuations in the demand and supply of each glycerol product in the international market; the
cost of glycerol transportation (the integration of glycerol fermentation processes into biodiesel
plants would be highly beneficial); the strict maintenance of sterility in the fermentation process,
and the risks involved in the downstream processing.
There are other limitations related to biotechnological conversions per se, such as the
usually slow kinetics of glycerol conversion (Khanna et al., 2012). Other drawbacks are related
to the low product concentration in the fermentation broth, the co-production of low-value by-
products (Cheng et al., 2012). To overcome these and other disadvantages, significant research
has been conducted to optimize the conditions and configurations of these processes. The most
commonly reported strategies are related to the isolation of novel glycerol-consuming strains and
the development of genetically-engineered strains (Khanna et al., 2012; Mattam et al., 2013).
However, breakthrough approaches related to the fermentation conditions applied throughout the
process have rarely been tested. For instance, these glycerol-based fermentations were still not
performed under non-conventional conditions, which could introduce significant improvements
in these processes, potentially making them more attractive for industrial production.
33
2.6. References
Ashby, R.D., Solaiman, D.K.Y., Foglia, T.A., 2005. Synthesis of short-/medium-chain-length
poly(hydroxyalkanoate) blends by mixed culture fermentation of glycerol.
Biomacromolecules 6, 2106–2112.
Ashby, R.D., Solaiman, D.K.Y., Foglia, T.A., 2004. Bacterial poly(hydroxyalkanoate) polymer
production from the biodiesel co-product stream. J. Polym. Environ. 12, 105–112.
Baeza-Jiménez, R., Lopez-Martinez, L.X., la Cruz-Medina, J., Espinosa-De-Los-Monteros, J.J.,
Garcia-Galindo, H.S., 2011. Effect of glucose on 1,3-propanediol production by
Lactobacillus reuteri. Rev. Mex. Ing. Química 10, 39–46.
Barbirato, F., Chedaille, D., Bories, A., 1997. Propionic acid fermentation from glycerol:
Comparison with conventional substrates. Appl. Microbiol. Biotechnol. 47, 441–446.
Beijerinck, M.W., Minkman, D.C.J., 1910. Bildung und verbrauch von stickoxydul durch
bakterien. Zentbl. Bakteriol. Parasitenkd. Infekt. Hyg. Abt. II 25, 30–63.
Biebl, H., 2001. Fermentation of glycerol by Clostridium pasteurianum - Batch and continuous
culture studies. J. Ind. Microbiol. Biotechnol. 27, 18–26.
Biebl, H., Zeng, A.-P., Menzel, K., Deckwer, W.-D., 1998. Fermentation of glycerol to 1,3-
propanediol and 2,3-butanediol by Klebsiella pneumoniae. Appl. Microbiol. Biotechnol. 50,
24–29.
Boenigk, R., Bowien, S., Gottschalk, G., 1993. Fermentation of glycerol to 1, 3-propanediol in
continuous cultures of Citrobacter freundii. Appl. Microbiol. Biotechnol. 38, 453–457.
Bormann, E.J., Roth, M., 1999. The production of polyhydroxybutyrate by Methylobacterium
rhodesianum and Ralstonia eutropha in media containing glycerol and casein hydrolysates.
Biotechnol. Lett. 21, 1059–1063.
Brämer, C.O., Vandamme, P., da Silva, L.F., Gomez, J.G., Steinbüchel, A., 2001.
Polyhydroxyalkanoate-accumulating bacterium isolated from soil of a sugar-cane plantation
in Brazil. Int. J. Syst. Evol. Microbiol. 51, 1709–1713.
Brigham, C.J., Sinskey, A.J., 2012. Applications of polyhydroxyalkanoates in the medical
industry. Int. J. Biotechnol. Wellness Ind. 1, 52–60.
Burgé, G., Saulou-Bérion, C., Moussa, M., Pollet, B., Flourat, A., Allais, F., Athès, V., Spinnler,
H.-E., 2015. Diversity of Lactobacillus reuteri strains in converting glycerol into 3-
hydroxypropionic acid. Appl. Biochem. Biotechnol. 177, 923–939.
Carvalho, M., Matos, M., Roca, C., Reis, M.A.M., 2014. Succinic acid production from glycerol
by Actinobacillus succinogenes using dimethylsulfoxide as electron acceptor. N.
Chen, G.-Q., Wu, Q., 2005. The application of polyhydroxyalkanoates as tissue engineering
materials. Biomaterials 26, 6565–6578.
34
Chen, X., Zhang, D.-J., Qi, W.-T., Gao, S.-J., Xiu, Z.-L., Xu, P., 2003. Microbial fed-batch
production of 1,3-propanediol by Klebsiella pneumoniae under micro-aerobic conditions.
Cheng, K.-K., Zhao, X.-B., Zeng, J., Zhang, J.-A., 2012. Biotechnological production of succinic
acid: current state and perspectives. Biofuels, Bioprod. Biorefining 6, 302–318.
Choi, J., Lee, S.Y., 1999. Factors affecting the economics of polyhydroxyalkanoate production
by bacterial fermentation. Appl. Microbiol. Biotechnol. 51, 13–21.
Ciriminna, R., Pina, C. Della, Rossi, M., Pagliaro, M., 2014. Understanding the glycerol market.
Eur. J. Lipid Sci. Technol. 116, 1432–1439.
da Cruz Pradella, J.G., Taciro, M.K., Mateus, A.Y.P., 2010. High-cell-density poly (3-
hydroxybutyrate) production from sucrose using Burkholderia sacchari culture in airlift
bioreactor. Bioresour. Technol. 101, 8355–8360.
da Silva, G.P., Mack, M., Contiero, J., 2009. Glycerol: A promising and abundant carbon source
for industrial microbiology. Biotechnol. Adv. 27, 30–39.
Darbon, E., Ito, K., Huang, H.-S., Yoshimoto, T., Poncet, S., Deutscher, J., 1999. Glycerol
transport and phosphoenolpyruvate-dependent enzyme I-and HPr-catalysed
phosphorylation of glycerol kinase in Thermus flavus. Microbiology 145, 3205–3212.
Deguchi, S., Shimoshige, H., Tsudome, M., Mukai, S., Corkery, R.W., Ito, S., Horikoshi, K.,
2011. Microbial growth at hyperaccelerations up to 403,627 x g. Proc. Natl. Acad. Sci. 108,
7997–8002.
Dharmadi, Y., Murarka, A., Gonzalez, R., 2006. Anaerobic fermentation of glycerol by
Escherichia coli: A new platform for metabolic engineering. Biotechnol. Bioeng. 94, 821–
829.
Dishisha, T., Pereyra, L.P., Pyo, S.-H., Britton, R.A., Hatti-Kaul, R., 2014. Flux analysis of the
Lactobacillus reuteri propanediol-utilization pathway for production of 3-
hydroxypropionaldehyde, 3-hydroxypropionic acid and 1,3-propanediol from glycerol.
Microb. Cell Fact. 13, 76.
Doleyres, Y., Beck, P., Vollenweider, S., Lacroix, C., 2005. Production of 3-
hydroxypropionaldehyde using a two-step process with Lactobacillus reuteri. Appl.
Microbiol. Biotechnol. 68, 467–474.
Dugar, D., Stephanopoulos, G., 2011. Relative potential of biosynthetic pathways for biofuels and
bio-based products. Nat. Biotechnol. 29, 1074–1078.
DuPont Tate & Lyle Bio Products, 2018. Zemea® USP-FCC Propanediol for Food and Flavors
[WWW Document]. URL http://www.duponttateandlyle.com/zemea_usp/food_and_flavors
(accessed 3.15.18).
Durgapal, M., Kumar, V., Yang, T.H., Lee, H.J., Seung, D., Park, S., 2014. Production of 1,3-
35
propanediol from glycerol using the newly isolated Klebsiella pneumoniae J2B. Bioresour.
Technol. 159, 223–231.
El-Ziney, M.G., Arneborg, N., Uyttendaele, M., Debevere, J., Jakobsen, M., 1998.
Characterization of growth and metabolite production of Lactobacillus reuteri during
glucose/glycerol cofermentation in batch and continuous cultures. Biotechnol. Lett. 20,
913–916.
Fu, D., Libson, A., Miercke, L.J.W., Weitzman, C., Nollert, P., Krucinski, J., Stroud, R.M., 2000.
Structure of a glycerol-conducting channel and the basis for its selectivity. Science (80-. ).
290, 481–486.
Giannopoulos, G., Sullivan, M.J., Hartop, K.R., Rowley, G., Gates, A.J., Watmough, N.J.,
Richardson, D.J., 2017. Tuning the modular Paracoccus denitrificans respirome to adapt
from aerobic respiration to anaerobic denitrification. Environ. Microbiol. 19, 4953–4964.
Grahame, D.A.S., Kang, T.S., Khan, N.H., Tanaka, T., 2013. Alkaline conditions stimulate the
production of 1,3-propanediol in Lactobacillus panis PM1 through shifting metabolic
pathways. World J. Microbiol. Biotechnol. 29, 1207–1215.
Heller, K.B., Lin, E.C., Wilson, T.H., 1980. Substrate specificity and transport properties of the
glycerol facilitator of Escherichia coli. J. Bacteriol. 144, 274–278.
Heyndrickx, M., De Vos, P., Vancanneyt, M., De Ley, J., 1991. The fermentation of glycerol by
Clostridium butyricum LMG 1212t2 and 1213t1 and C. pasteurianum LMG 3285. Appl.
Hong, A.-A., Cheng, K.-K., Peng, F., Zhou, S., Sun, Y., Liu, C.-M., Liu, D.-H., 2009. Strain
isolation and optimization of process parameters for bioconversion of glycerol to lactic acid.
J. Chem. Technol. Biotechnol. 84, 1576–1581.
Huijberts, G.N., Eggink, G., De Waard, P., Huisman, G.W., Witholt, B., 1992. Pseudomonas
putida KT2442 cultivated on glucose accumulates poly(3-hydroxyalkanoates) consisting of
saturated and unsaturated monomers. Appl. Environ. Microbiol. 58, 536–544.
Ibrahim, M.H.A., Steinbüchel, A., 2010. Zobellella denitrificans strain MW1, a newly isolated
bacterium suitable for poly (3-hydroxybutyrate) production from glycerol. J. Appl.
Microbiol. 108, 214–225.
Iino, T., Uchimura, T., Komagata, K., 2002. The effect of sodium acetate on the growth yield, the
production of L-and D-lactic acid, and the activity of some enzymes of the glycolytic
pathway of Lactobacillus sakei NRIC 1071T and Lactobacillus plantarum NRIC 1067T. J.
Gen. Appl. Microbiol. 48, 91–102.
Ji, X.-J., Huang, H., Zhu, J.-G., Hu, N., Li, S., 2009. Efficient 1,3-propanediol production by fed-
batch culture of Klebsiella pneumoniae: The role of pH fluctuation. Appl. Biochem.
36
Johnson, D.T., Taconi, K.A., 2007. The glycerin glut: Options for the value-added conversion of
crude glycerol resulting from biodiesel production. Environ. Prog. Sustain. Energy 26, 338–
348.
Johnson, E., Sarchami, T., Kießlich, S., Munch, G., Rehmann, L., 2016. Consolidating biofuel
platforms through the fermentative bioconversion of crude glycerol to butanol. World J.
Microbiol. Biotechnol. 32, 103.
Jolly, J., Hitzmann, B., Ramalingam, S., Ramachandran, K.B., 2014. Biosynthesis of 1,3-
propanediol from glycerol with Lactobacillus reuteri: Effect of operating variables. J.
Biosci. Bioeng. 118, 188–194.
Kalaiyezhini, D., Ramachandran, K.B., 2015. Biosynthesis of poly-3-hydroxybutyrate (PHB)
from glycerol by Paracoccus denitrificans in a batch bioreactor: Effect of process variables.
Prep. Biochem. Biotechnol. 45, 69–83.
Katryniok, B., Paul, S., Dumeignil, F., 2013. Recent developments in the field of catalytic
dehydration of glycerol to acrolein. ACS Catal. 3, 1819–1834.
Kaur, G., Srivastava, A.K., Chand, S., 2012. Advances in biotechnological production of 1,3-
propanediol. Biochem. Eng. J. 64, 106–118.
Khanna, S., Goyal, A., Moholkar, V.S., 2012. Microbial conversion of glycerol: Present status
and future prospects. Crit. Rev. Biotechnol. 32, 235–262.
Koller, M., Marsalek, L., 2015. Potential of diverse prokaryotic organisms for glycerol-based
polyhydroxyalkanoate production. Appl. Food Biotechnol. 2, 3–15.
Kumar, P., Jun, H.-B., Kim, B.S., 2018. Co-production of polyhydroxyalkanoates and carotenoids
through bioconversion of glycerol by Paracoccus sp. strain LL1. Int. J. Biol. Macromol.
107, 2552–2558.
Kumar, V., Ashok, S., Park, S., 2013. Recent advances in biological production of 3-
hydroxypropionic acid. Biotechnol. Adv. 31, 945–961.
Kumar, V., Park, S., 2018. Potential and limitations of Klebsiella pneumoniae as a microbial cell
factory utilizing glycerol as the carbon source. Biotechnol. Adv. 36, 150–167.
Lee, C.S., Aroua, M.K., Daud, W.M.A.W., Cognet, P., Pérès-Lucchese, Y., Fabre, P.L., Reynes,
O., Latapie, L., 2015. A review: Conversion of bioglycerol into 1,3-propanediol via
biological and chemical method. Renew. Sustain. Energy Rev. 42, 963–972.
Lee, P.C., Lee, W.G., Lee, S.Y., Chang, H.N., 2001. Succinic acid production with reduced by-
product formation in the fermentation of Anaerobiospirillum succiniciproducens using
glycerol as a carbon source. Biotechnol. Bioeng. 72, 41–48.
Lee, S.Y., 1996. Bacterial polyhydroxyalkanoates. Biotechnol. Bioeng. 49, 1–14.
Lin, R., Liu, H., Hao, J., Cheng, K., Liu, D., 2005. Enhancement of 1,3-propanediol production
by Klebsiella pneumoniae with fumarate addition. Biotechnol. Lett. 27, 1755–1759.
37
Lindlbauer, K.A., Marx, H., Sauer, M., 2017. Effect of carbon pulsing on the redox household of
Lactobacillus diolivorans in order to enhance 1,3-propanediol production. N. Biotechnol.
34, 32–39.
Liu, H.-J., Zhang, D.-J., Xu, Y.-H., Mu, Y., Sun, Y.-Q., Xiu, Z.-L., 2007. Microbial production
of 1,3-propanediol from glycerol by Klebsiella pneumoniae under micro-aerobic conditions
up to a pilot scale. Biotechnol. Lett. 29, 1281–1285.
Lu, J., Tappel, R.C., Nomura, C.T., 2009. Mini-review: Biosynthesis of poly(hydroxyalkanoates).
J. Macromol. Sci. Part C Polym. Rev. 49, 226–248.
Luo, L.H., Seo, J.-W., Baek, J.-O., Oh, B.-R., Heo, S.-Y., Hong, W.-K., Kim, D.-H., Kim, C.H.,
2011. Identification and characterization of the propanediol utilization protein PduP of
Lactobacillus reuteri for 3-hydroxypropionic acid production from glycerol. Appl.
Lüthi-Peng, Q., Dileme, F., Puhan, Z., 2002. Effect of glucose on glycerol bioconversion by
Lactobacillus reuteri. Appl. Microbiol. Biotechnol. 59, 289–296.
Madison, L.L., Huisman, G.W., 1999. Metabolic engineering of poly (3-hydroxyalkanoates):
From DNA to plastic. Microbiol. Mol. Biol. Rev. 63, 21–53.
Maervoet, V.E.T., De Mey, M., Beauprez, J., De Maeseneire, S., Soetaert, W.K., 2010. Enhancing
the microbial conversion of glycerol to 1,3-propanediol using metabolic engineering. Org.
Process Res. Dev. 15, 189–202.
Mattam, A.J., Clomburg, J.M., Gonzalez, R., Yazdani, S.S., 2013. Fermentation of glycerol and
production of valuable chemical and biofuel molecules. Biotechnol. Lett. 35, 831–842.
McNutt, J., Yang, J., others, 2017. Utilization of the residual glycerol from biodiesel production
for renewable energy generation. Renew. Sustain. Energy Rev. 71, 63–76.
Menzel, K., Zeng, A.-P., Deckwer, W.-D., 1997. High concentration and productivity of 1,3-
propanediol from continuous fermentation of glycerol by Klebsiella pneumoniae. Enzyme
Microb. Technol. 20, 82–86.
Metsoviti, M., Paraskevaidi, K., Koutinas, A., Zeng, A.-P., Papanikolaou, S., 2012. Production of
1,3-propanediol, 2,3-butanediol and ethanol by a newly isolated Klebsiella oxytoca strain
growing on biodiesel-derived glycerol based media. Process Biochem. 47, 1872–1882.
Mitrea, L., Trif, M., Cuatoi, A.-F., Vodnar, D.-C., 2017. Utilization of biodiesel derived-glycerol
for 1, 3-PD and citric acid production. Microb. Cell Fact. 16, 190.
Mothes, G., Schnorpfeil, C., Ackermann, J.-U., 2007. Production of PHB from crude glycerol.
Eng. Life Sci. 7, 475–479.
Możejko-Ciesielska, J., Kiewisz, R., 2016. Bacterial polyhydroxyalkanoates: Still fabulous?
Microbiol. Res. 192, 271–282.
Murakami, N., Oba, M., Iwamoto, M., Tashiro, Y., Noguchi, T., Bonkohara, K., Abdel-Rahman,
38
M.A., Zendo, T., Shimoda, M., Sakai, K., others, 2016. L-Lactic acid production from
glycerol coupled with acetic acid metabolism by Enterococcus faecalis without carbon loss.
J. Biosci. Bioeng. 121, 89–95.
OECD, 2011. Biofuels, in OECD-FAO Agricultural Outlook 2011–2020. Paris.
Peplinski, K., Ehrenreich, A., Döring, C., Bömeke, M., Reinecke, F., Hutmacher, C., Steinbüchel,
A., 2010. Genome-wide transcriptome analyses of the ‘Knallgas’ bacterium Ralstonia
eutropha H16 with regard to polyhydroxyalkanoate metabolism. Microbiology 156, 2136–
2152.
Petitdemange, E., Dürr, C., Andaloussi, S.A., Raval, G., 1995. Fermentation of raw glycerol to
1,3-propanediol by new strains of Clostridium butyricum. J. Ind. Microbiol. Biotechnol. 15,
498–502.
Pflügl, S., Marx, H., Mattanovich, D., Sauer, M., 2012. 1, 3-Propanediol production from glycerol
with Lactobacillus diolivorans. Bioresour. Technol. 119, 133–140.
Poli, A., Di Donato, P., Abbamondi, G.R., Nicolaus, B., 2011. Synthesis, production, and
biotechnological applications of exopolysaccharides and polyhydroxyalkanoates by
archaea. Archaea 2011, 693253.
Prada-Palomo, Y., Romero-Vanegas, M., Díaz-Ruíz, P., Molina-Velasco, D., Guzmán-Luna, C.,
2012. Lactic acid production by Lactobacillus sp. from biodiesel derived raw glycerol.
CT&F - Ciencia, Tecnol. y Futur. 5, 57–65.
Przystałowska, H., Lipiński, D., Słomski, R., others, 2015. Biotechnological conversion of
glycerol from biofuels to 1,3-propanediol using Escherichia coli. Acta Biochim. Pol. 62,
23–34.
Ragout, A., Sineriz, F., Diekmann, H., De Valdez, G.F., 1996. Shifts in the fermentation balance
of Lactobacillus reuteri in the presence of glycerol. Biotechnol. Lett. 18, 1105–1108.
Raynaud, C., Sarçabal, P., Meynial-Salles, I., Croux, C., Soucaille, P., 2003. Molecular
characterization of the 1,3-propanediol (1,3-PD) operon of Clostridium butyricum. Proc.
Natl. Acad. Sci. 100, 5010–5015.
Research TM, 2012. 1,3-Propanediol market: Global industry analysis, size, share, growth, trends
and forecasts 2013–2019.
Ricci, M.A., Russo, A., Pisano, I., Palmieri, L., de Angelis, M., Agrimi, G., 2015. Improved 1,3-
propanediol synthesis from glycerol by the robust Lactobacillus reuteri strain DSM 20016.
J. Microbioliology Biotechnol. 25, 893–902.
Rodríguez-Contreras, A., Koller, M., Miranda-de Sousa Dias, M., Calafell-Monfort, M.,
Braunegg, G., Marqués-Calvo, M.S., 2015. Influence of glycerol on poly(3-
hydroxybutyrate) production by Cupriavidus necator and Burkholderia sacchari. Biochem.
Eng. J. 94, 50–57.
39
Rodriguez, A., Wojtusik, M., Ripoll, V., Santos, V.E., Garcia-Ochoa, F., 2016. 1,3-Propanediol
production from glycerol with a novel biocatalyst Shimwellia blattae ATCC 33430:
Operational conditions and kinetics in batch cultivations. Bioresour. Technol. 200, 830–837.
Rothermich, M.M., Guerrero, R., Lenz, R.W., Goodwin, S., 2000. Characterization, seasonal
occurrence, and diel fluctuation of poly(hydroxyalkanoate) in photosynthetic microbial
mats. Appl. Environ. Microbiol. 66, 4279–4291.
Rywińska, A., Rymowicz, W., 2010. High-yield production of citric acid by Yarrowia lipolytica
on glycerol in repeated-batch bioreactors. J. Ind. Microbiol. Biotechnol. 37, 431–435.
Sabet-Azad, R., Linares-Pastén, J.A., Torkelson, L., Sardari, R.R.R., Hatti-Kaul, R., 2013.
Coenzyme A-acylating propionaldehyde dehydrogenase (PduP) from Lactobacillus reuteri:
Kinetic characterization and molecular modeling. Enzyme Microb. Technol. 53, 235–242.
San Kong, P., Aroua, M.K., Daud, W.M.A.W., 2016. Conversion of crude and pure glycerol into
derivatives: a feasibility evaluation. Renew. Sustain. Energy Rev. 63, 533–555.
Sardari, R.R.R., Dishisha, T., Pyo, S.-H., Hatti-Kaul, R., 2013a. Biotransformation of glycerol to
3-hydroxypropionaldehyde: improved production by in situ complexation with bisulfite in
a fed-batch mode and separation on anion exchanger. J. Biotechnol. 168, 534–542.
Sardari, R.R.R., Dishisha, T., Pyo, S.-H., Hatti-Kaul, R., 2013b. Improved production of 3-
hydroxypropionaldehyde by complex formation with bisulfite during biotransformation of
glycerol. Biotechnol. Bioeng. 110, 1243–1248.
Sauer, M., Russmayer, H., Grabherr, R., Peterbauer, C.K., Marx, H., 2017. The efficient clade:
Lactic acid bacteria for industrial chemical production. Trends Biotechnol. 35, 756–769.
Scholten, E., Renz, T., Thomas, J., 2009. Continuous cultivation approach for fermentative
succinic acid production from crude glycerol by Basfia succiniciproducens DD1.
Biotechnol. Lett. 31, 1947–1951.
Schütz, H., Radler, F., 1984. Anaerobic reduction of glycerol to propanediol-1,3 by Lactobacillus
brevis and Lactobacillus buchneri. Syst. Appl. Microbiol. 5, 169–178.
Shahid, S., Mosrati, R., Ledauphin, J., Amiel, C., Fontaine, P., Gaillard, J.-L., Corroler, D., 2013.
Impact of carbon source and variable nitrogen conditions on bacterial biosynthesis of
polyhydroxyalkanoates: evidence of an atypical metabolism in Bacillus megaterium DSM
509. J. Biosci. Bioeng. 116, 302–308.
Shimizu, R., Chou, K., Orita, I., Suzuki, Y., Nakamura, S., Fukui, T., 2013. Detection of phase-
dependent transcriptomic changes and Rubisco-mediated CO2 fixation into poly(3-
hydroxybutyrate) under heterotrophic condition in Ralstonia eutropha H16 based on RNA-
seq and gene deletion analyses. BMC Microbiol. 13, 169.
Sobolov, M., Smiley, K.L., 1960. Metabolism of glycerol by an acrolein-forming Lactobacillus.
J. Bacteriol. 79, 261–266.
40
Sriramulu, D.D., Liang, M., Hernandez-Romero, D., Raux-Deery, E., Lünsdorf, H., Parsons, J.B.,
Warren, M.J., Prentice, M.B., 2008. Lactobacillus reuteri DSM 20016 produces cobalamin-
dependent diol dehydratase in metabolosomes and metabolizes 1,2-propanediol by
disproportionation. J. Bacteriol. 190, 4559–4567.
Sun, D., Yamada, Y., Sato, S., Ueda, W., 2017. Glycerol as a potential renewable raw material
for acrylic acid production. Green Chem. 19, 3186–3213.
Suriyamongkol, P., Weselake, R., Narine, S., Moloney, M., Shah, S., 2007. Biotechnological
approaches for the production of polyhydroxyalkanoates in microorganisms and plants - A
review. Biotechnol. Adv. 25, 148–175.
Syu, M.-J., 2001. Biological production of 2,3-butanediol. Appl. Microbiol. Biotechnol. 55, 10–
18.
Talarico, T.L., Dobrogosz, W.J., 1990. Purification and characterization of glycerol dehydratase
from Lactobacillus reuteri. Appl. Environ. Microbiol. 56, 1195–1197.
Talarico, T.L., Dobrogosz, W.J., 1989. Chemical characterization of an antimicrobial substance
produced by Lactobacillus reuteri. Antimicrob. Agents Chemother. 33, 674–679.
Tan, G.-Y.A., Chen, C.-L., Li, L., Ge, L., Wang, L., Razaad, I.M.N., Li, Y., Zhao, L., Mo, Y.,
Wang, J.-Y., 2014. Start a research on biopolymer polyhydroxyalkanoate (PHA): A review.
Polymers (Basel). 6, 706–754.
Tanadchangsaeng, N., Yu, J., 2012. Microbial synthesis of polyhydroxybutyrate from glycerol:
Gluconeogenesis, molecular weight and material properties of biopolyester. Biotechnol.
Bioeng. 109, 2808–2818.
Thomson, A.J., Giannopoulos, G., Pretty, J., Baggs, E.M., Richardson, D.J., 2012. Biological
sources and sinks of nitrous oxide and strategies to mitigate emissions. Philos. Trans. R.
Soc. B Biol. Sci. 367, 1157–1168.
Traub, H.L., Hirt, P., Herlinger, H., 1995. Mechanical properties of fibers made of
polytrimethylene terephthalate. Chem. Fibers Int. 45, 110–111.
Ueda, S., Matsumoto, S., Takagi, A., Yamane, T., 1992. Synthesis of poly(3-hydroxybutyrate-co-
3-hydroxyvalerate) from methanol and n-amyl alcohol by the methylotrophic bacteria
Paracoccus denitrificans and Methylobacterium extorquens. Appl. Environ. Microbiol. 58,
3574–3579.
Verlinden, R.A.J., Hill, D.J., Kenward, M.A., Williams, C.D., Radecka, I., 2007. Bacterial
synthesis of biodegradable polyhydroxyalkanoates. J. Appl. Microbiol. 102, 1437–1449.
Vieira, P.B., Kilikian, B. V, Bastos, R. V, Perpetuo, E.A., Nascimento, C.A.O., 2015. Process
strategies for enhanced production of 1,3-propanediol by Lactobacillus reuteri using
glycerol as a co-substrate. Biochem. Eng. J. 94, 30–38.
Vivek, N., Pandey, A., Binod, P., 2016. Biological valorization of pure and crude glycerol into
41
1,3-propanediol using a novel isolate Lactobacillus brevis N1E9. 3.3. Bioresour. Technol.
213, 222–230.
Vivek, N., Sindhu, R., Madhavan, A., Anju, A.J., Castro, E., Faraco, V., Pandey, A., Binod, P.,
2017. Recent advances in the production of value added chemicals and lipids utilizing
biodiesel industry generated crude glycerol as a substrate - Metabolic aspects, challenges
and possibilities: An overview. Bioresour. Technol. 239, 507–517.
Voegele, R.T., Sweet, G.D., Boos, W., 1993. Glycerol kinase of Escherichia coli is activated by
interaction with the glycerol facilitator. J. Bacteriol. 175, 1087–1094.
Wang, Z., Ammar, E.M., Zhang, A., Wang, L., Lin, M., Yang, S.-T., 2015. Engineering
Propionibacterium freudenreichii subsp. shermanii for enhanced propionic acid
fermentation: Effects of overexpressing propionyl-CoA: Succinate CoA transferase. Metab.
Eng. 27, 46–56.
Wilkens, E., Ringel, A.K., Hortig, D., Willke, T., Vorlop, K.-D., 2012. High-level production of
1, 3-propanediol from crude glycerol by Clostridium butyricum AKR102a. Appl. Microbiol.
Biotechnol. 93, 1057–1063.
Williams, S.F., Martin, D.P., Horowitz, D.M., Peoples, O.P., 1999. PHA applications: Addressing
the price performance issue: I. Tissue engineering. Int. J. Biol. Macromol. 25, 111–121.
Yamane, T., 1993. Yield of poly-D(-)-3-hydroxybutyrate from various carbon sources: A
theoretical study. Biotechnol. Bioeng. 41, 165–170.
Yamane, T., Chen, X.-F., Ueda, S., 1996a. Polyhydroxyalkanoate synthesis from alcohols during
the growth of Paracoccus denitrificans. FEMS Microbiol. Lett. 135, 207–211.
Yamane, T., Chen, X., Ueda, S., 1996b. Growth-associated production of poly(3-
hydroxyvalerate) from n-pentanol by a methylotrophic bacterium, Paracoccus denitrificans.
Yang, G., Tian, J., Li, J., 2007. Fermentation of 1,3-propanediol by a lactate deficient mutant of
Klebsiella oxytoca under microaerobic conditions. Appl. Microbiol. Biotechnol. 73, 1017–
1024.
Yang, T.-W., Rao, Z.-M., Zhang, X., Xu, M.-J., Xu, Z.-H., Yang, S.-T., 2013. Fermentation of
biodiesel-derived glycerol by Bacillus amyloliquefaciens: Effects of co-substrates on 2,3-
butanediol production. Appl. Microbiol. Biotechnol. 97, 7651–7658.
Yazdani, S.S., Gonzalez, R., 2007. Anaerobic fermentation of glycerol: a path to economic
viability for the biofuels industry. Curr. Opin. Biotechnol. 18, 213–219.
Zaushitsyna, O., Dishisha, T., Hatti-Kaul, R., Mattiasson, B., 2017. Crosslinked, cryostructured
Lactobacillus reuteri monoliths for production of 3-hydroxypropionaldehyde, 3-
hydroxypropionic acid and 1, 3-propanediol from glycerol. J. Biotechnol. 241, 22–32.
Zeng, A.-P., Biebl, H., 2002. Bulk-chemicals from biotechnology: The case of 1,3-propanediol
42
production and the new trends. Adv. Biochem. Eng. Biotechnol. 74, 239–259.
Zhao, L., Ma, X., Zheng, Y., Zhang, J., Wei, G., Wei, D., 2009. Over-expression of glycerol
dehydrogenase and 1,3-propanediol oxidoreductase in Klebsiella pneumoniae and their
effects on conversion of glycerol into 1, 3-propanediol in resting cell system. J. Chem.
Technol. Biotechnol. 84, 626–632.
Zhu, Y., Li, J., Tan, M., Liu, L., Jiang, L., Sun, J., Lee, P., Du, G., Chen, J., 2010. Optimization
and scale-up of propionic acid production by propionic acid-tolerant Propionibacterium
acidipropionici with glycerol as the carbon source. Bioresour. Technol. 101, 8902–8906.
Zinn, M., Witholt, B., Egli, T., 2001. Occurrence, synthesis and medical application of bacterial
polyhydroxyalkanoate. Adv. Drug Deliv. Rev. 53, 5–21.
43
44
CHAPTER III
Scope and outline of the work

CHAPTER III. Scope and outline of the work
The microbial conversion of glycerol to value-added products constitutes a promising

approach to dispose the excess glycerol generated by the biodiesel industry. However, there are
still some limitations related to these biotechnological processes, such as the usually slow kinetics
of glycerol conversion, the low product concentration, and the co-production of low-value by-
products. Different strategies have been tested to overcome these issues. One possibility is the
application of non-conventional conditions (such as HP, electric fields, or ultrasounds) to
glycerol-based fermentations, possibly introducing significant improvements in these processes.
Therefore, the main goal of the present work was to evaluate the application of sub-lethal HP on
fermentative processes that use glycerol as substrate to the production of value-added products.
The work intended to i) understand the possible effects of pressure on glycerol fermentations,
providing fundamental knowledge about the microbial responses to pressure as a variable of life;
and ii) if possible, contribute to the process improvement, by increasing fermentation rates,
increasing the yields of desired products, and/or promoting the formation of different products.
In order to perform this study, two specific microbial processes were selected: one of
them with Lactobacillus reuteri, which converts glycerol into 1,3-PDO; and the other with
Paracoccus denitrificans, which is able to convert glycerol into PHA. These processes have
different oxygen requirements, since the former corresponds to an anaerobic process, and the
latter corresponds to an aerobic one. The combination of both was intentionally selected to allow
a more comprehensive perspective regarding the potential of HP for application on glycerol-based
processes. In both cases, only sub-lethal HP conditions (in the range of 10 to 50 MPa) were used,
at controlled temperature, with proper fermentative media, and using glycerol as substrate (or co-
substrate, in the case of L. reuteri).
The following chapters describe the results obtained for each of these glycerol-based
processes. The first part of the study (Chapters IV – VI) is focused on the glycerol/glucose co-
fermentation by L. reuteri. Chapter IV approaches the general effects of HP on this process, using
different initial acetate concentrations in the culture medium, as a possible microbial growth
inhibitor. Different HP levels (10, 25 and 35 MPa) were applied to L. reuteri fermentation, and
the effects on cell growth, substrate consumption, and product formation were evaluated over
time. In Chapter V, adaptation of L. reuteri to pressure was assessed during four consecutive
fermentation cycles under sub-lethal HP levels (10 and 25 MPa). Similarly to the previous
chapter, different initial acetate concentrations in the culture medium were tested. To evaluate the
impact of pressure throughout the fermentation cycles, cell growth, substrate consumption and
product formation were analyzed after each cycle. In addition, the effects of pressure were
evaluated on cell membrane permeability, by determining the leakage of nucleic acids and
extracellular proteins to the extracellular medium. To better understand the effects of these HP-
cycles on the L. reuteri metabolome, the metabolic profiles of these samples were characterized
47
CHAPTER III. Scope and outline of the work
by 1H NMR spectroscopy coupled with multivariate analysis, to evaluate the changes in the
abundances of several metabolites relatively to atmospheric pressure (Chapter VI).
The work regarding glycerol fermentation by P. denitrificans is approached in Chapters
VII and VIII. In this particular case, it was necessary to give special focus on air/oxygen
availability, since this is a critical parameter that affects P. denitrificans growth and metabolic
profile. It is important to highlight that the HP system used in the present work involves some
process constrains, related to limited air volumes and the absence of agitation. Accordingly, the
study presented in Chapter VII was divided in two main goals: i) to study the effects of air
availability on P. denitrificans; ii) to assess if the strain was able to grow and maintain metabolic
activity under HP (10, 25 and 35 MPa). Those aspects were evaluated in terms of biomass
concentration, substrate consumption and formation of extracellular products. A more detailed
analysis of the HP effects (10, 25, 35 and 50 MPa) on P. denitrificans is presented in Chapter
VIII, which includes the evaluation of cell growth and substrate formation and PHA formation,
as well as characterization of the PHA monomeric composition at different pressure conditions.
48
CHAPTER IV
Lactobacillus reuteri growth and

fermentation under high pressure towards
the production of 1,3-propanediol
Adapted from:
Mota, M.J., Lopes, R.P., Sousa, S., Gomes, A.M., Delgadillo, I., Saraiva, J.A., Lactobacillus
reuteri growth and fermentation under high pressure towards the production of 1,3-propanediol.
Submitted in Food Research International.
CHAPTER IV. L. reuteri growth and fermentation under high pressure
4.1. Introduction
Biodiesel production has been raising some sustainability issues, and one of them
corresponds to the formation of crude glycerol as a reaction by-product (da Silva et al., 2009;
Kolesárová et al., 2011). Glycerol may be used as substrate in microbial fermentation processes,
for production of high-value products, such as organic acids, alcohols, or biopolymers (Mattam
et al., 2013). In this context, Lactobacillus reuteri presents itself as a lactic acid bacteria (LAB)
with high potential for the production of different relevant compounds from glycerol, including
reuterin, 3-hydroxypropionic acid and 1,3-propanediol (1,3-PDO) (Burgé et al., 2015; Dishisha
et al., 2015, 2014; Jolly et al., 2014; Lee et al., 2015; Ricci et al., 2015; Vollenweider and Lacroix,
2004). 1,3-Propanediol is an important chemical intermediate used in food, cosmetic and
pharmaceutical industries (Biebl et al., 1999; Katrlík et al., 2007; Menzel et al., 1997; Saxena et
al., 2009). It is also a GRAS (Generally Recognized As Safe) food and beverage ingredient,
commercialized by DuPont Tate & Lyle Bio Products. It can be used as a flavor carrier, to reduce
bitterness and enhance sweetness, or as a humectant, or processing aid. It is approved for use in
different food products, such as seasonings and flavorings, confections and frostings, nuts and nut
products, alcoholic beverages, and frozen dairy products (DuPont Tate & Lyle Bio Products,
2018). The global market of 1,3-PDO has considerably increased in recent years, due to the
increasing market demand of its derivatives – e.g. 60,000 tons in 2012 versus 150,000 tons
expected to 2019 (Lee et al., 2015; Research TM, 2012).
The production of 1,3-PDO by L. reuteri from glycerol seems a promising biotechnological
approach. For that, a co-fermentation with glycerol and a sugar is usually performed, since this
LAB uses glycerol only as an electron acceptor and not as a carbon source for growth (Ricci et
al., 2015). For instance, Ragout et al. (1996) showed that the production of 1,3-PDO by L. reuteri
only occurred during the maltose/glycerol co-fermentation. An important factor for this
fermentation is the presence of acetate, which has been recognized as a microbial growth
inhibitor. However, Iino et al. (2002) have stated that certain acetate concentrations may have a
positive influence on growth and fermentation of LAB. A possible explanation is the ability of
acetate to act as an indirect proton and electron acceptor, leading to the activation of oxidative
pathways at early stages of fermentation (Heyndrickx et al., 1991; Jolly et al., 2014). In fact, Jolly
et al. (2014) reported that an initial acetate concentration of 5 g L-1 increased 1,3-PDO
productivity, while concentrations above this value were found to be inhibitory.
The performance of fermentation under non-conventional conditions (such as high
pressure, electric fields or ultrasounds) is a strategy that has been recently tested for different
fermentation processes, aiming the stimulation of microbial growth and/or improvement of
fermentation rates and yields (Mota et al., 2018). This approach involves the use of these
technologies, at sub-lethal levels, prior and/or during the fermentation process, in order to affect
51
cell growth and metabolism, but without compromising cell viability. In some cases, considerable
improvements were achieved, such as increased yields, productivities and fermentation rates,
lower accumulation of by-products and/or production of different compounds.
High pressure (HP) is particularly advantageous in the context of microbial fermentations,
since it can be maintained continuously during the entire fermentation process, without heat
generation and no need of energy to maintain the pressure (Mota et al., 2018). The beneficial
effects of HP on bioprocesses have been reported for several bacteria and yeasts (Mota et al.,
2013). Picard et al. (2007) reported acceleration of alcoholic fermentation by Saccharomyces
cerevisiae at 5 and 10 MPa, with the production of ethanol proceeding 3-fold faster at 10 MPa
compared to atmospheric pressure. High pressure can also be used to change the metabolic
selectivity of the fermentative strains. For instance, application of pressures of 7 and 17 MPa
during fermentation by Clostridium thermocellum redirected the metabolism from the production
of by-products (such as acetate) to ethanol, leading to an increase in the ratio ethanol:acetate up
to 60-fold compared to atmospheric pressure (Bothun et al., 2004). Another example is the
utilization of HP to modify the properties of biopolymers produced during fermentation.
Production of bacterial cellulose by Gluconacetobacter xylinus under HP (30, 60 and 100 MPa)
showed profound differences in morphological properties of the polymer, having a significantly
higher density compared with the cellulose produced at atmospheric pressure (Kato et al., 2007).
The effects of HP were already tested during fermentation of a food product: pressures up to 100
MPa were applied during lactic acid fermentation for probiotic yogurt production. High pressure
reduced the fermentation rate, but it was still possible to produce yogurt at 5 MPa, showing that
lactic acid fermentation can be performed at this pressure without compromising the fermentation
activity of the bacterial strains (including the probiotic one) (Mota et al., 2015).
Based on the results published so far, application of HP seems to be a promising approach
that could introduce significant improvements in glycerol-based fermentation processes.
Therefore, the present work intended to study the effects of HP (10 - 35 MPa) on growth and
fermentation of L. reuteri DSM 20016, and using different initial acetate concentrations in the
culture medium, to test its possible microbial growth inhibitor effect under pressure.
4.2. Material and methods
4.2.1. Microorganism and culture media

A lyophilized culture of Lactobacillus reuteri DSM 20016 obtained from DSMZ,
Germany, was used in this study. The strain was reconstituted on commercial MRS medium,
52
according to the manufacturer’s instructions. Cells were maintained at -80 ºC in a cryoprotectant

solution.
Commercial MRS broth was used for seed culture preparation and growth. For the
fermentation experiments, two different MRS broth media were used, one with initial acetate (5
g L-1) and the other without it.
4.2.2. Seed culture preparation

A single cell colony was seeded into 10 ml MRS broth and incubated for 10-12 h at 37 ºC,
in static conditions. The culture was then transferred to a 250 ml modified MRS medium in a 300
ml Erlenmeyer flask, incubated overnight at 37 ºC, in static conditions.
4.2.3. Fermentation experiments

The medium was inoculated with 10 % (v/v) of seed culture. The mixture was homogenized
and then transferred to polyethylene bags, which were carefully heat-sealed with no air. All these
steps were performed in an aseptic environment, within a laminar flow cabinet, to avoid sample
contamination.
Fermentation was carried at 37 ºC under different HP conditions (10, 25, and 35), for 32 h.
All fermentations were performed under unaerated conditions. These experiments were
conducted in a Hydrostatic press (FPG7100, Stanstead Fluid Power, Stanstead, United Kingdom),
with a pressure vessel of 100 mm inner diameter and 250 mm height surrounded by an external
jacket to control the temperature, using a mixture of propylene glycol and water as pressurizing
fluid. In parallel, a control sample carried out fermentation at atmospheric pressure (0.1 MPa),
while maintaining constant the remaining process conditions. At all pressure conditions
(including atmospheric pressure), two different types of samples were studied: i) with acetate,
corresponding to samples with an initial acetate concentration of 5 g L-1; ii) and without acetate,
corresponding to samples with no acetate added in the culture medium. In all cases, fermentation
experiments were performed in duplicate and the analyses were also performed in duplicate.
4.2.4. Determination of biomass concentration

Biomass concentration of the samples was determined by optical density measurement at
600 nm, with a Multiskan GO Microplate Spectrophotometer (Thermo Fisher Scientific Inc.,
USA). Cell dry weight (CDW) was routinely determined using a standard curve relating L. reuteri
optical density and cell dry weight (CDW).
53
4.2.5. Determination of viable cell counts

For determination of viable cells, serial dilutions (using Ringer solution) of the culture
samples were prepared and aliquots of 1.0 mL of proper dilutions were plated in MRS agar plates,
incubated at 37 °C for 24 h.
4.2.6. Quantification of glycerol, glucose, and fermentation products

Culture samples were centrifuged at 10,000 rpm and 4 ºC for 10 min and the collected
supernatants were filtered through a 0.22 μm filter membrane. Analysis by HPLC was performed
using a HPLC Knauer system equipped with Knauer K-2301 RI and K-2501 UV detectors, and
an Aminex HPX-87H cation exchange column (300 x 7.8 mm) (Bio-Rad Laboratories Pty Ltd,
Hercules, CA, USA). The mobile phase was 13 mM H 2SO4, delivered at a flow rate of 0.6 mL
min-1 and the column maintained at 65 °C. Peaks were identified by their retention times and
quantified using calibration curves prepared with different standards. Because of co-elution of
glycerol and lactate, and since glycerol is only detected by the RI detector, but lactate is detected
by both detectors, lactate concentration was estimated through the UV detector. The RI area
corresponding to this lactate concentration was estimated using the corresponding RI calibration
curve. Glycerol quantification was determined by the difference between the total area of the RI
peak and the calculated contribution of lactate in the area of the peak.
4.2.7. Statistical analysis

The results obtained for the previously indicated parameters were tested at a 0.05 level of
probability and the effect of pressure was tested with a one-way analysis of variance (ANOVA),
followed by a multiple comparisons test (Tukey HSD) to identify the differences between
samples.
4.3. Results and Discussion
4.3.1. Effect of sub-lethal pressure on cell growth

In order to understand how sub-lethal pressures affect L. reuteri growth, biomass
concentration and viable cell counts were both determined. Figure 4.1 shows the variation of both
parameters during fermentation at different pressure conditions (0.1, 10, 25, and 35 MPa), for the
samples with acetate in the culture medium (Figures 4.1.a and 4.1.b), and the ones without added
acetate (Figures 4.1.c and 4.1.d).
The presence of acetate in the culture medium was found to affect biomass production and
cell viability, which confirms that this is an important aspect to consider in L. reuteri fermentation
54
processes. At atmospheric pressure (0.1 MPa), samples without acetate showed higher cell counts
and slightly higher biomass concentrations, compared do the ones with acetate, which may result
from an inhibitory effect caused by accumulation of this metabolite: the acetate produced from
glucose metabolism accumulates, in addition to the amount already present in the medium and,
in consequence, the inhibitory concentration is more rapidly achieved. The effect of acetate on
microbial growth is a highly debated subject on literature, since it is usually regarded as an
inhibitor of microbial growth (Lasko et al., 2000), but some studies stated it may have a favorable
influence on growth rates and energy yields of lactic acid bacteria (Iino et al., 2002; Jolly et al.,
2014).
Figure 4.1. Biomass concentration and viable cell counts throughout fermentation, for samples with initial
acetate (a and b) and without initial acetate (c and d), according to the pressure applied: 10 MPa (triangles,
▲), 25 MPa (diamonds ♦), or 35 MPa (stars *). Control samples (0.1 MPa) are represented as squares (■).
Application of HP was also shown to affect biomass concentration during L. reuteri

fermentation. In general, HP decreased biomass concentration (Figures 4.1.a and 4.1.c), with a
more pronounced effect at higher pressure levels. However, the presence of acetate in the culture
media seems to increase the L. reuteri tolerance to HP. In the samples with acetate, biomass
concentrations showed similar behavior at 0.1 and 10 MPa, reaching similar (p > 0.05)
concentrations (2.36 and 2.43 g L-1, respectively) after 32 h. When fermentation was carried out
55
without acetate in the culture medium, the increase in biomass concentration was slower at 10
MPa than at 0.1 MPa, but it reached similar (p > 0.05) values after 32 h (2.84 at 0.1 MPa, and
2.68 g L-1 at 10 MPa). In samples with acetate fermenting at 25 MPa, variation in biomass
concentration was minimal during the first 24 h, which suggests a longer lag phase for
development of appropriate stress responses to ensure cell survival and growth at this pressure
level. After that time, biomass concentration increased up to 1.72 g L-1 at 32 h of fermentation,
which is still considerably lower than the concentrations achieved at 0.1 and 10 MPa (p < 0.05).
However, it is important to note that samples without acetate showed no variation in biomass
concentration at 25 MPa, indicating higher sensitivity to this pressure in the absence of acetate.
At 35 MPa, the pressure seems to severely affect L. reuteri cells, which were not able to withstand
this level of pressure and, in consequence, biomass concentration remained constant over the
fermentation time, regardless of the presence/absence of acetate in the medium.
Viable cell counts (Figures 4.1.b and 4.1.d) were also affected by HP, with the effect
depending on the pressure level and the acetate in the medium. In the samples with acetate, cell
counts were higher (p < 0.05) at 10 MPa than at 0.1 MPa, and even higher (p < 0.05) at 25 MPa.
Similarly to the observed for biomass concentration, viable cell counts reflected higher sensitivity
of L. reuteri to pressure in the absence of acetate. In this case, cell counts were still similar at 0.1
MPa and 10 MPa, but considerably lower (p < 0.05) at 25 MPa. These higher cell counts at 25
MPa are not in accordance with the biomass concentration results, which were lower at this
pressure compared to 10 and 0.1 MPa. This may be explained by the longer lag phase at 25 MPa
that lasted for the first 24 h of fermentation. After that time, cells at 10 and 0.1 MPa were already
reaching the stationary phase, due to accumulation of acetate and other inhibitory metabolites,
which resulted in a subsequent reduction of cell viability. In contrast, cells at 25 MPa were still
growing at that time, promoting the increase of viable cell counts. In both cases, there was no cell
growth at 35 MPa, suggesting microbial inhibition or even inactivation after 24 h, followed by a
slight increase, possibly as a result of development of HP stress responses. This indicates a
considerably longer lag phase when fermentation was carried out at 35 MPa.
The results in this section showed that fermentation under pressures between 10 and 35
MPa affected L. reuteri growth, with the effects being dependent on the initial acetate
concentration. In fact, the presence of acetate seems to enhance the resistance of this strain to
pressure. Although there is no explanation for this behavior on literature, this effect suggests that
acetate may have a role on the cell response mechanisms developed under HP-stress, possibly due
to cross-protection mechanisms. It is documented in literature that LAB response to HP can be
related to other factors, such as heat or cold, and their ability to react to pressure can be explained
by a bacteria cross-protection system (Bucka-Kolendo and Sokołowska, 2017; Ljungh and
56
Wadström, 2009; Scheyhing et al., 2004). However, these LAB response to HP is still not well
documented and understood, and must be clarified by further transcriptome and proteome studies.
4.3.2. Effect of sub-lethal pressure on substrate consumption

The consumption of glycerol and glucose was monitored during fermentation at different
pressure conditions, and the results are presented in Figure 4.2. In all cases, the uptake of glycerol
(Figure 4.2.a and 4.2.c) and glucose (Figure 4.2.b and 4.2.d) seem to be correlated, with both
showing similar behavior, independently of the pressure applied or the initial concentration of
acetate. Although glycerol and glucose are metabolized by different pathways, these are, in fact,
dependent of each other. The regulation of glycerol metabolizing pathway is dependent on the
availability of glucose (Lüthi-Peng et al., 2002), since the NADH generated from glucose by the
glycolysis pathway is recycled by the presence of glycerol and its transformation into 1,3-PDO
(Doleyres et al., 2005). Therefore, high glucose consumption usually results in high glycerol
uptake rate (Jolly et al., 2014).
Figure 4.2. Glycerol and glucose concentrations throughout fermentation, for samples with initial acetate
(a and b) and without initial acetate (c and d), according to the pressure applied: 10 MPa (triangles, ▲), 25
MPa (diamonds ♦), or 35 MPa (stars *). Control samples (0.1 MPa) are represented as squares (■).
57
The presence/absence of acetate was found to affect substrate consumption under different
pressure conditions. With acetate (Figures 4.2.a and 4.2.b), fermentation at 0.1 and 10 MPa
showed similar substrate consumption profiles over time. After 32 h, samples at 10 MPa reached
glycerol and glucose concentrations of 6.37 and 2.02 g L-1, respectively, which were similar (p >
0.05) to the obtained at 0.1 MPa (6.51 and 1.96 g L-1 for glycerol and glucose, respectively). These
final concentrations correspond to consumption of ≈ 63 % of the glycerol and ≈ 83 % of the
glucose present in the medium (Table 4.1). At 25 MPa, the lag phase was more extensive (as
discussed in the previous section), with low substrate consumption during the first 24 h. After that
time, the consumption of both substrates had a considerable increase, resulting in a glycerol
consumption of ≈ 30 % and a glucose consumption of ≈ 42 % after 32 h (Table 4.1), which are
still lower (p < 0.05) than the values for samples at 0.1 and 10 MPa. However, the extension of
fermentation time at 25 MPa would possibly lead to substrate consumptions similar to the
observed at 0.1 and MPa, since the L. reuteri cells seem to be able to grow and ferment at this
pressure, suggesting the development of suitable response mechanisms that allowed the
microorganism to withstand these stressful pressure conditions. Differently, application of 35
MPa promoted low substrate consumption during the entire fermentation time: glycerol
concentration had no variation over time, while glucose concentration showed a slight decrease
(correspondent to consumption of 6.28 % of the total glucose present in the medium).
Table 4.1. Consumption (%) of glycerol and glucose after 32 h of fermentation at different pressure
conditions, in samples with and without acetate.
Samples Pressure Glycerol consumed (%) Glucose consumed (%)
0.1 MPa 61.96 83.54

10 MPa 62.76 83.05
With acetate
25 MPa 29.99 41.73
35 MPa - 6.28
0.1 MPa 61.85 89.72
10 MPa 50.85 75.61
Without acetate
25 MPa - -
35 MPa 2.35 -
For the samples without added acetate (Figures 4.2.c and 4.2.d), substrate consumption was
more affected by the pressure increase. In this case, glycerol and glucose consumptions at 10 MPa
(50.85 and 75.61 %, respectively, after 32 h) were significantly lower (p < 0.05) than at 0.1 MPa
(61.85 and 89.72 %, respectively, after 32 h). At 25 MPa, there was no consumption of either
58
glycerol or glucose during the 32 h, possibly due to the inhibition of L. reuteri metabolism.
However, cell growth (i.e. increase of viable cell counts) was still observed in these samples. In
contrast, L. reuteri was able to consume glycerol (≈ 2 % after 32 h) when fermentation was carried
out at 35 MPa, but the viable cell counts after 24 h and 32 h were lower than in the beginning of
the process. Considering these results, it is not clear if the HP effects at 25 and 35 MPa promote
complete inhibition of cell growth and metabolism, but this aspect will be clarified in the
following sections.
4.3.3. Effect of sub-lethal pressure on production of 1,3-propanediol

The production of 1,3-PDO during fermentation at different pressure conditions was also
evaluated, with the results being presented in Figure 4.3. In addition, 1,3-PDO yields on glycerol
(Y1,3-PDO/Gly, g g-1) and 1,3-PDO productivities (Q1,3-PDO, g L-1 h-1) are represented in Table 4.2.
Figure 4.3. 1,3-Propanediol concentration throughout fermentation, for samples with initial acetate (a) and
without initial acetate (b), according to the pressure applied: 10 MPa (triangles, ▲), 25 MPa (diamonds ♦),
or 35 MPa (stars *). Control samples (0.1 MPa) are represented as squares (■).
At atmospheric pressure, L. reuteri reached 1,3-PDO titers of 3.76 and 3.30 g L-1 after 32
h, which are in the range of some of the values reported in literature for L. reuteri glycerol/sugar
co-fermentation (≈ 4 g L-1) (El-Ziney et al., 1998; Ragout et al., 1996). However, these are
considerably lower than some of the values reported in literature for fermentation with LAB. For
instance, Lactobacillus diolivorans has been reported to produce 1,3-PDO concentrations as high
as 74 g L-1 or 92 g L-1 in fed-batch experiments (Lindlbauer et al., 2017; Pflügl et al., 2012). For
fermentation with L. reuteri, Jolly et al. (2014) reached 1,3-PDO concentrations of 65.3 g L-1
during fed-batch experiments, while Baeza-Jiménez et al. (2011) and Ricci et al. (2015) obtained
concentrations of 29 and 41 g L-1, respectively, for batch cultivations. Nevertheless, it is important
to note that all of these studies were performed in bioreactors, with monitoring, adjustment and
59
optimization of several important process parameters. In the present work, fermentation was
carried out with a different and simpler experimental set-up, with lower volumes suitable for the
HP experiments, as a proof of concept to evaluate if it was possible to carry out the process under
these pressure conditions, and if it could promote any improvement.
Table 4.2. 1,3-PDO yields on glycerol (Y1,3-PDO/Gly) and 1,3-PDO productivities (Q1,3-PDO), for fermentation
at different pressure conditions, in samples with and without acetate.
Samples Pressure Y1,3-PDO/Gly (g g-1) Q1,3-PDO (g L-1 h-1)
0.1 MPa 0.354 ± 0.009 0.117 ± 0.002

10 MPa 0.392 ± 0.005 0.131 ± 0.001
With acetate
25 MPa 0.355 ± 0.022 0.057 ± 0.006
35 MPa - 0.002 ± 0.000
0.1 MPa 0.328 ± 0.028 0.103 ± 0.004
10 MPa 0.414 ± 0.031 0.107 ± 0.004
Without acetate
25 MPa - -
35 MPa - -
Yields and productivities were calculated from a single time-point corresponding to the end of the experiment (32 h). Values reported
in the table represent the mean ± SD of two independent biological replicates.
As indicated in Figure 4.3, the presence/absence of acetate affected the production of 1,3-
PDO under different pressure conditions. In general, higher rates and titers were achieved in the
samples with acetate (Figure 4.3.a), compared to the ones without this compound in the culture
medium (Figure 4.3.b). Interestingly, the opposite effect was noted for cell growth at atmospheric
pressure, since the samples without acetate had higher cell counts and biomass concentrations,
compared do the ones with acetate. This suggests that although acetate acts as an inhibitor of L.
reuteri growth, it has a positive effect on the production of 1,3-PDO. Jolly et al. (2014) published
similar findings, with an initial acetate concentration up to 5 g L-1 improving 1,3-PDO
productivity. However, the authors also reported an improvement of L. reuteri cell growth
promoted by acetate, which was not observed in the present work.
The presence of acetate in the culture media seems to increase the L. reuteri tolerance to
HP, resulting in an enhancement of 1,3-PDO production at these conditions. The production of
this compound was faster at 10 MPa and resulted in a higher (p < 0.05) final titer (4.21 g L-1),
when compared to 0.1 MPa (3.76 g L-1). In consequence, fermentation at 10 MPa showed yield
and productivity improvements of 11 and 12 %, respectively, relatively to 0.1 MPa. By increasing
the pressure to higher levels, 1,3-PDO production tends to decrease. At 25 MPa, this compound
60
was only detected after 24 h of fermentation, which is in accordance to the observed for cell
growth and substrate consumption. Therefore, the final titer at this pressure (1.83 g L -1) is
significantly lower (p < 0.05) than at 10 and 0.1 MPa, which results in a decreased productivity
of 1,3-PDO. In contrast, the yield reached at 25 MPa is similar to the obtained at 0.1 MPa (0.354
and 0.355 g g-1, respectively), due to an also lower glycerol consumption at 25 MPa. When
fermentation was carried at 35 MPa, the production of 1,3-PDO was minimal, indicating that this
pressure might be enough to inhibit L. reuteri metabolism from glycerol, at least up to the
fermentation time studied (32 h).
In samples without initial acetate, 1,3-PDO production was only observed at 0.1 and 10
MPa. At 10 MPa, 1,3-PDO was slower than at 0.1 MPa, resulting in lower concentrations after
24 and 28 h. However, the final titers after 32 h of fermentation were similar (p > 0.05) for both
conditions (3.30 and 3.42 g L-1, at 0.1 and 10 MPa, respectively), and even slightly higher at 10
MPa. The 1,3-PDO productivity was also similar at 0.1 and 10 MPa (0.103 and 0.107 g L -1 h-1,
respectively), while the yield was ≈ 26 % higher at 10 MPa. This improved yield is a result of
similar 1,3-PDO titers obtained with a lower glycerol consumption, which suggests that a higher
proportion of the consumed glycerol was being directed to the production of 1,3-PDO.
It can be concluded that, under specific conditions, application of pressure (10 MPa) on
fermentation may stimulate production of 1,3-PDO by L. reuteri, promoting enhanced final titers,
yields and productivities. Picard et al. (2007) reported a similar effect on alcoholic fermentation,
when applying pressure in the range of 5-10 MPa to Saccharomyces cerevisiae. In that case, the
production of bioethanol at 10 MPa proceeded 3-fold faster than at 0.1 MPa, and the fermentation
yield was increased by 6 % and 5 %, at 5 and 10 MPa, respectively. However, fermentation was
slowed down at pressures above 20 MPa. Despite the use of a different microorganism in a
different fermentation process, these results are in accordance to the obtained in the present work.
The mechanisms behind the increase of fermentative activity by HP are still not completely
understood. Several effects of HP on microbial cells are related to modification of cell membrane
composition and fluidity, which affect the exchange of substances (both substrates and
metabolites) throughout the cell, with consequent effect on cell metabolism (Mota et al., 2013).
Other factors may also be involved, such as the effects of HP on enzyme activity (Eisenmenger
and Reyes-De-Corcuera, 2009) and on regulation of gene expression (Bravim et al., 2012;
Fernandes et al., 2004). Picard et al. (2007) indicated some possible explanations for the
improvement of bioethanol production by S. cerevisiae. These included the increased activity of
one or more enzymes involved in the fermentation pathways; the improvement of ethanol
secretion from the cell, leading to a decreased intracellular concentration and to the reduction of
retro-inhibition; and the enhancement of substrate consumption. The last one. However, is not
valid for the present work, since substrate consumption was evaluated and it was demonstrated
61
that HP did not promote the increase of glycerol uptake. Therefore, the improvements in 1,3-PDO
production may be related to changes in cell membrane, gene expression and enzymatic activity.
4.3.4. Effect of sub-lethal pressure on formation of fermentation by-products

During glycerol/glucose co-fermentation, glucose enters into the glycolysis pathway,
leading to the formation of different products, such as ethanol, lactate, acetate, and others. The
presence of glycerol and its transformation into 1,3-PDO enables the cells to recycle the NADH
generated during glycolysis (Doleyres et al., 2005). However, some of these products obtained
from the glycolysis pathway will compete with 3-HPA (the precursor of 1,3-PDO) for NADH-
oxidoreductase and, as a result, its formation may decrease the 1,3-PDO yield (El-Ziney et al.,
1998). The production of these by-products is addressed in this section. Figure 4.4 shows the
production of lactate (Figures 4.4.a and 4.4.d), acetate (Figures 4.4.b and 4.4.e) and ethanol
(Figures 4.4.c and 4.4.f) during fermentation at different pressure conditions, while the respective
yields on glucose (Y, g g-1) are represented in Table 4.3.
Lactate was the major by-product of L. reuteri fermentation process, reaching final titers
as high as 7.56 g L-1 at 0.1 MPa, in samples without initial acetate. Lactate is an interesting
chemical building block and hence its formation could represent a valuable feature of this process
(Ricci et al., 2015), even if it is not the intended in the present work. However, it is also important
to consider that accumulation of lactate has an inhibitory effect on fermentation, due to its high
toxicity levels (Ricci et al., 2015; Vieira et al., 2015).
At atmospheric pressure, the absence of acetate in the initial culture medium stimulated the
production of lactate, leading to a higher final titer compared to the samples with initial acetate
(7.56 and 6.77 g L-1, respectively). In general, application of HP seems to inhibit lactate
production, with this effect being more pronounced in the samples without acetate. However, in
the presence of acetate (Figure 4.4.a), the production of lactate was faster at 10 MPa than at 0.1
MPa. By increasing the pressure to 25 and 35 MPa, lactate production over time was reduced, but
still detected at both pressures (2.77 and 0.35 g L-1, respectively).In the absence of acetate (Figure
4.4.d), application of HP at 10 MPa considerably decreased (p < 0.05) lactate production, leading
to a final titer of 3.77 g L-1, while no lactate production was detected at 25 MPa. At 35 MPa, a
negligible concentration lactate (0.04 g L-1) was obtained. Therefore, application of HP on L.
reuteri fermentation had a general inhibitory effect on lactate production, except at 10 MPa when
acetate was added in the culture medium. A similar inhibitory effect was reported in a previous
work using three different LAB (Streptococcus thermophilus, Lactobacillus bulgaricus and
Bifidobacterium lactis) for probiotic yogurt production. In that study, application of HP in the
range of 5 and 100 MPa negatively affected cell growth and lactic acid fermentation (Mota et al.,
2015). However, in the particular case of the present work, lactate is not the desired metabolic
62
product, and its production can even (indirectly) reduce the yields of 1,3-PDO. In addition, lactate
is the fermentation product with the highest toxicity for L. reuteri (Ricci et al., 2015), with a
concentration of 4 g L-1 being enough to inhibit cell growth (Vieira et al., 2015). Considering all
those aspects, the reduction of lactate production under some HP conditions may represent a
potential process improvement.
Figure 4.4. Lactate, acetate and ethanol concentrations throughout fermentation, for samples with initial
acetate (a, b and c) and without initial acetate (c, d and e), according to the pressure applied: 10 MPa
(triangles, ▲), 25 MPa (diamonds ♦), or 35 MPa (stars *). Control samples (0.1 MPa) are represented as
squares (■).
63
Table 4.3. Yields of lactate (YLact/Glu), acetate (YAcet/Glu), and ethanol (YEtOH/Glu) on glucose, for fermentation
at different pressure conditions, in samples with and without acetate.
Samples Pressure YLact/Glu (g g-1) YAcet/Glu (g g-1) YEtOH/Glu (g g-1)
0.1 MPa 0.682 ± 0.004 0.287 ± 0.010 0.119 ± 0.002

10 MPa 0.706 ± 0.064 0.277 ± 0.006 0.103 ± 0.001
With acetate
25 MPa 0.557 ± 0.011 0.332 ± 0.031 0.108 ± 0.007
35 MPa 0.469 ± 0.011 - -
0.1 MPa 0.709 ± 0.002 0.333 ± 0.005 0.088 ± 0.001
10 MPa 0.419 ± 0.021 0.172 ± 0.001 0.059 ± 0.003
Without acetate
25 MPa - - -
35 MPa - - -
Yields were calculated from a single time-point corresponding to the end of the experiment (32 h). Values reported in the table
represent the mean ± SD of two independent biological replicates.
Acetate is another by-product of L. reuteri fermentation, which has a relevant role on the
regulation of this process. The presence/absence of initial acetate in the culture medium was found
to affect its production during fermentation. As expected, the final titer was higher in samples
with acetate added in the medium, but higher acetate production was observed in the samples
without initial acetate (≈ 3.6 g L-1 of acetate produced, compared to ≈ 2.9 g L-1 in samples without
acetate). The increase of pressure promoted a general decrease in acetate production during
fermentation. With initial acetate (Figure 4.4.b), fermentation at 0.1 and 10 MPa showed close
acetate concentration profiles over time, with similar (p > 0.05) final titers of 5.84 and 5.72 g L -
1
, respectively. At 25 MPa, acetate concentration only started to increase after 24 h, achieving a
final titer of 4.64 g L-1. In this case, acetate yield was higher than at 0.1 and 10 MPa (with an
increase of 12 % and 20 %, respectively), suggesting a stimulation of acetate formation at 25
MPa. In samples without acetate (Figure 4.4.e), acetate production was considerably reduced
when fermentation was performed at 10 and 25 MPa, leading to lower (p < 0.05) final titers (1.75
and 0.87 g L-1) compared to 0.1 MPa (3.76 g L-1). As a consequence, the acetate yield at 10 MPa
was nearly 2-fold lower than the yield at 0.1 MPa, which demonstrates a clear inhibitory effect of
HP on acetate production in these samples. At 35 MPa, no acetate production was detected during
fermentation, regardless of the initial acetate concentration.
Another product that can be obtained from L. reuteri glucose metabolism is ethanol, but
usually at low concentrations in glycerol/glucose co-fermentation processes (Jolly et al., 2014;
Ricci et al., 2015). In fact, the highest ethanol concentration in this study corresponded to 1.19 g
L-1, which is considerably lower than the concentrations of other fermentation by-products.
64
Ethanol production was even lower under HP, regardless of the presence or absence of acetate in
the initial culture medium. In samples with acetate (Figure 4.4.c), ethanol production was detected
not only at 0.1 MPa, but also at 10 and 25 MPa. In the case of samples without acetate (Figure
4.4.f), ethanol was only produced at 0.1 and 10 MPa, with considerably lower (p < 0.05)
concentrations at 10 MPa.
Overall, the results in this section suggest an inhibitory effect of HP on the formation of
lactate, acetate and ethanol. In order to obtain an optimum yield of 1,3-PDO, these compounds
should be maintained at the lowest level or even completely stopped (Lee et al., 2015). It is
important to note that these by-products and 1,3-PDO are produced from different substrates using
different metabolic pathways, but both are necessarily related, due to NADH/NAD + recycling and
competition. Because of that, regulation of glycerol bioconversion is highly dependent on glucose
metabolism. In addition, acetate, ethanol, and mostly lactate may act as inhibitors of L. reuteri
growth and fermentation (Ricci et al., 2015; Vieira et al., 2015). Because of all that, 1,3-PDO
biosynthetic ability and yield may be improved by minimizing by-product formation (Jolly et al.,
2014; Lüthi-Peng et al., 2002).
To better understand the effects of HP on L. reuteri metabolic selectivity, molar ratios
between 1,3-PDO and the by-products were estimated for all pressure conditions, in the presence
and absence of initial acetate (Table 4.4).
Table 4.4. Molar ratios between 1,3-PDO and by-products (lactate, acetate, and ethanol) produced after 32
h of fermentation at different pressure conditions, in samples with and without acetate.
Molar ratio
Samples Pressure
1,3-PDO:by-products
0.1 MPa 0.249 ± 0.001
10 MPa 0.284 ± 0.013
With acetate
25 MPa 0.200 ± 0.010
35 MPa 0.014 ± 0.000
0.1 MPa 0.260 ± 0.009
10 MPa 0.546 ± 0.009
Without acetate
25 MPa -
35 MPa -
Regardless of the presence/absence of acetate, the highest 1,3-PDO:by-products ratios were

achieved at 10 MPa, indicating that L. reuteri product selectivity has shifted towards production
of 1,3-PDO at this pressure. This effect was considerably more pronounced when fermentation
65
was carried out without initial acetate: in this case, the ratio at 10 MPa has more than doubled
relatively to 0.1 MPa (0.546 and 0.260, respectively). Since the highest 1,3-PDO yield (0.414 g
g-1) was also reached at these fermentation conditions, these seem to be the most suitable to
perform L. reuteri fermentation for 1,3-PDO production. By increasing pressure to higher
pressure levels (25 and 35 MPa), the 1,3-PDO/by-products ratios decreased, or were even
impossible to estimate.
Modification of product selectivity by application of HP at 10 MPa would represent a
valuable improvement of the process, not only due to enhancement of 1,3-PDO yields, but also
because of the formation of undesired by-products that may limit the scale-up of fermentation and
its widespread application in the industry. The mechanisms promoting these HP effects are still
not clearly explained in literature. Bothun et al. (2004) observed shifts in product selectivity
towards ethanol production by Clostridium thermocellum under HP (7 and 17 MPa), and
explained it with possible mass-action effects and changes in membrane fluidity. Both of these
aspects may also be related to the modifications in product selectivity observed in the present
work, even if concerning a different microorganism and a different fermentation process.
Although the impact of HP on microbial cells vary according to the microbial strain in question,
pressure exerts some general effects on living organisms, predominantly affecting some key
pressure sensitive structures and processes (Mota et al., 2013).
4.4. Conclusions
This corresponds to the first study regarding the effects of HP on L. reuteri, or on any
glycerol fermentation process. High pressure (in the range of 10 and 35 MPa) was found to affect
L. reuteri growth and co-fermentation, with the effects varying not only according to the applied
pressure level, but also to the initial acetate content in the samples. In general, fermentation was
less inhibited by HP when acetate was present in the initial culture medium, indicating that acetate
enhances the resistance of L. reuteri to pressure. This suggests that acetate may have a role on the
cell response mechanisms developed under HP stress.
Under specific conditions, application of HP on L. reuteri co-fermentation stimulated 1,3-
PDO production. At 10 MPa, higher titers, yields and productivities were observed, compared to
the same process at 0.1 MPa. This enhancement effect was more pronounced in the presence of
acetate in the culture medium: in this case, 1,3-PDO production was faster at 10 MPa, while yield
and productivity were increased by 11 and 12 %, respectively, relatively to 0.1 MPa. In samples
without initial acetate, 1,3-PDO production was slower at 10 MPa than at 0.1 MPa, but similar
titers were obtained after 32 h, and the yield was ≈ 26 % higher at 10 MPa. In any case, the 1,3-
PDO concentrations achieved in this work were lower compared to some studies in literature.
66
However, similarly to other studies of fermentation under pressure, we intended to perform a

study in a small scale, as a proof of concept to assess possible improvements in microbial growth
and fermentation. Further process optimization and scale-up could lead to enhanced production
of 1,3-PDO, with concentrations in the range of those reported for fermentation in bioreactors.
Overall, HP can be a useful tool to modify and improve L. reuteri fermentation, by using
specific conditions adapted to that purpose. Moreover, this opens the way for utilization of this
technology to stimulate other fermentations processes, resulting in a real impact of HP technology
in the food and biotechnological fields.
67
4.5. References
Biebl, H., Menzel, K., Zeng, A.-P., Deckwer, W.-D., 1999. Microbial production of 1,3-
propanediol. Appl. Microbiol. Biotechnol. 52, 289–297.
Bucka-Kolendo, J., Sokołowska, B., 2017. Lactic acid bacteria stress response to preservation
processes in the beverage and juice industry. Acta Biochim. Pol. 64, 459–464.
Dishisha, T., Pyo, S.-H., Hatti-Kaul, R., 2015. Bio-based 3-hydroxypropionic-and acrylic acid
production from biodiesel glycerol via integrated microbial and chemical catalysis. Microb.
Cell Fact. 14, 200.
(accessed 3.15.18).
Eisenmenger, M.J., Reyes-De-Corcuera, J.I., 2009. High pressure enhancement of enzymes: A
review. Enzyme Microb. Technol. 45, 331–347.
68

913–916.
Fernandes, P.M.B., Domitrovic, T., Kao, C.M., Kurtenbach, E., 2004. Genomic expression
pattern in Saccharomyces cerevisiae cells in response to high hydrostatic pressure. Febs
Lett. 556, 153–160.
698.
Katrlík, J., Voštiar, I., Šefčovičová, J., Tkáč, J., Mastihuba, V., Valach, M., Štefuca, V.,
Gemeiner, P., 2007. A novel microbial biosensor based on cells of Gluconobacter oxydans
for the selective determination of 1,3-propanediol in the presence of glycerol and its
application to bioprocess monitoring. Anal. Bioanal. Chem. 388, 287–295.
Kolesárová, N., Hutnan, M., Bodík, I., Špalková, V., 2011. Utilization of biodiesel by-products
for biogas production. Biomed Res. Int. 2011, 126798.
Lasko, D.R., Zamboni, N., Sauer, U., 2000. Bacterial response to acetate challenge: A comparison
of tolerance among species. Appl. Microbiol. Biotechnol. 54, 243–247.
34, 32–39.
Ljungh, Å., Wadström, T., 2009. Lactobacillus stress responses, in: Ljungh, Å., Wadström, T.
(Eds.), Lactobacillus Molecular Biology: From Genomic to Probiotics. Caister Academy
69
Press, Norfolk, U.K., pp. 115–129.

Saxena, R.K., Anand, P., Saran, S., Isar, J., 2009. Microbial production of 1,3-propanediol:
Recent developments and emerging opportunities. Biotechnol. Adv. 27, 895–913.
Scheyhing, C.H., Hörmann, S., Ehrmann, M.A., Vogel, R.F., 2004. Barotolerance is inducible by
preincubation under hydrostatic pressure, cold-, osmotic- and acid-stress conditions in
Lactobacillus sanfranciscensis DSM 20451T. Lett. Appl. Microbiol. 39, 284–289.
Vieira, P.B., Kilikian, B. V, Bastos, R. V, Perpetuo, E.A., Nascimento, C.A.O., 2015. Process
strategies for enhanced production of 1,3-propanediol by Lactobacillus reuteri using
glycerol as a co-substrate. Biochem. Eng. J. 94, 30–38.
70
Vollenweider, S., Lacroix, C., 2004. 3-Hydroxypropionaldehyde: Applications and perspectives

of biotechnological production. Appl. Microbiol. Biotechnol. 64, 16–27.
71
72
CHAPTER V
Utilization of glycerol during consecutive

cycles of Lactobacillus reuteri fermentation
under pressure: The impact on cell growth
and fermentation profile
Adapted from:
Mota, M.J., Lopes, R.P., Sousa, S., Gomes, A.M., Lorenzo, J.M., Barba, F.J., Delgadillo, I.,
Saraiva, J.A., Utilization of glycerol during consecutive cycles of Lactobacillus reuteri
fermentation under pressure: The impact on cell growth and fermentation profile. Submitted in
Process Biochemistry.
CHAPTER V. Utilization of glycerol during consecutive cycles of L. reuteri fermentation under pressure
5.1. Introduction
Lactobacillus reuteri is a lactic acid bacterium (LAB) with high biotechnological value,
since it is able to produce relevant biochemical compounds from glycerol, including 1,3-
propanediol (1,3-PDO) (Burgé et al., 2015; Dishisha et al., 2015, 2014; Jolly et al., 2014; Lee et
al., 2015; Ricci et al., 2015; Vollenweider and Lacroix, 2004). This compound is a chemical
intermediate, widely used in different industries (Biebl et al., 1999; Katrlík et al., 2007; Menzel
et al., 1997; Saxena et al., 2009). It is also a GRAS (Generally Recognized As Safe) food and
beverage ingredient, commercialized by DuPont Tate & Lyle Bio Products (DuPont Tate & Lyle
Bio Products, 2018). The global market of 1,3-PDO has considerably increased in recent years,
with the global demand achieving 60.2 ktons in 2012, and being expect to reach ≈ 150.0 ktons in
2019 (Lee et al., 2015; Research TM, 2012).
For the production of 1,3-PDO by L. reuteri, a co-fermentation with glycerol and glucose
is usually performed, since glycerol is used only as an electron acceptor, and not as a carbon
source for growth (Ricci et al., 2015). Glycerol metabolism in L. reuteri occurs in two steps: in
the first one, glycerol is dehydrated into 3-hydroxypropionaldehyde (3-HPA); and in the second
step, 3-HPA is reduced into 1,3-PDO (El-Ziney et al., 1998; Schütz and Radler, 1984; Talarico
and Dobrogosz, 1989). In contrast, glucose is metabolized by the glycolysis pathway, leading to
the formation of various by-products, such as ethanol, lactate, acetate, and others (El-Ziney et al.,
1998; Talarico and Dobrogosz, 1989). Although both substrates are metabolized by alternative
pathways, the production of 1,3-PDO from glycerol is dependent on glucose availability (Lüthi-
Peng et al., 2002), as well as other factors, such as aeration and acetate concentration (Jolly et al.,
2014). Acetate has been recognized as a microbial growth inhibitor, but some studies report a
positive influence of certain acetate concentrations on LAB growth and fermentation (Iino et al.,
2002; Jolly et al., 2014). A possible explanation is the ability of acetate to act as an indirect proton
and electron acceptor, leading to the activation of oxidative pathway at early stages of
fermentation (Heyndrickx et al., 1991; Jolly et al., 2014).
The concept of sub-lethal high pressure (HP) is gaining relevance in the last decade,
mainly due to piezophilic microorganisms, which are showing to have considerable interest for
use in biotechnology (Mota et al., 2013). For instance, piezophilic microorganisms living under
deep-sea environments are attracting the attention of several pharmaceutical companies wishing
to obtain sources of compounds with relevant biological activity and novel properties (Mota et
al., 2018). Piezophiles are not the only types of microorganisms with capacity to withstand high
pressure, since some mesophilic microorganisms develop mechanisms to improve pressure
resistance, which allow them to grow under these stress conditions (Mota et al., 2013; Oger and
Jebbar, 2010). The exposure of bacterial cells to stressful conditions, such as HP, during growth
and fermentation, involves a complex network of response mechanisms, with several metabolic
75
activities that will reflect upon the metabolome of the fermentative microorganisms, and thus on
the bioproducts and on the bioprocess itself (Serrazanetti et al., 2009). Some of the changes
promoted by these stress responses may have a positive outcome, such as improved fermentation
rates, yields and productivities (Bravim et al., 2012; Picard et al., 2007), changes in metabolic
selectivity (Bothun et al., 2004), or even production of compounds with different properties (Kato
et al., 2007). In this context, application of sub-lethal HP is particularly suitable and has high
versatility, as it can be applied as intermittent stresses or continuously during the whole
fermentation time, without serious cell loss and no heating effect. In addition, it is only necessary
to provide energy to generate the pressure (and not to maintain it) and so, application of HP stress
during the whole fermentation process has minimal energetic costs.
Since stress is one of the major driving forces of microbial evolution and adaptation
(Serrazanetti et al., 2009), application of sub-lethal HP may promote the development of
microorganisms with new features, some of them with potential technological interest.
Considering that, the present work intended to assess adaptation of L. reuteri DSM 20016 to
pressure, during four consecutive fermentation cycles under sub-lethal HP levels (10 and 25
MPa). Cell viability, growth and fermentation profiles were evaluated throughout the study, to
understand how pressure affected each fermentation cycle, giving special focus to the production
of 1,3-PDO.

A lyophilized culture of Lactobacillus reuteri DSM 20016 obtained from DSMZ,
Germany, was used in this study. The strain was reconstituted on commercial MRS medium,
according to the manufacturer’s instructions. Cells were maintained at -80 ºC in a cryoprotectant
solution.
Commercial MRS medium was used for seed culture preparation and growth. For the
fermentation experiments, two different media were used: i) with an initial acetate concentration
of 5 g L-1, correspondent to commercial MRS broth; and ii) without initial acetate, as described
by Jolly et al. (2014).

A single cell colony was seeded into 10 mL MRS broth and incubated for 10 - 12 h at 37
ºC, in static conditions. The culture was then transferred to a 250 mL MRS medium without
acetate in a 300 mL Erlenmeyer flask, and incubated overnight at 37 ºC, in static conditions.
76

The medium was inoculated with 10 % (v/v) of seed culture. The mixture was
homogenized and then transferred to polyethylene bags, which were carefully sealed with the
minimal air possible. All these steps were performed in an aseptic environment, within a laminar
flow cabinet, to avoid sample contamination.
Fermentation experiments were performed during four consecutive cycles of 24 h, at 37
ºC, under different pressure conditions (10 and 25 MPa). After each cycle, the fermentative
medium was collected, analyzed, and used as inoculum for the subsequent fermentation cycle,
carried out at the same pressure conditions (Figure 5.1). All fermentations were performed under
unaerated conditions. These experiments were conducted in a HP vessel SFP FPG7100 (Stansted
Fluid Power Ltd, Essex, UK), using a mixture of propylene glycol and water (40:60) as
pressurizing fluid. This equipment is a hydrostatic press with a pressure vessel of 100 mm inner
diameter and 250 mm height, surrounded by an external jacket to control the temperature. In
parallel, a control sample carried out fermentation at atmospheric pressure (0.1 MPa), while
maintaining constant the remaining process conditions. To all pressure conditions (including
atmospheric pressure), two different types of samples were studied: i) with acetate, corresponding
to samples with an initial acetate concentration of 5 g L-1; ii) and without acetate, corresponding
to samples with no acetate added in the culture medium. In all cases, fermentation experiments
were performed in duplicate. Samples were collected after each cycle, and all the analyses were
also performed in duplicate.
Figure 5.1. Schematic representation of the fermentation experiments, carried out in four consecutive
cycles of 24 h, at 10 or 25 MPa. Fermentation cycles were also performed at 0.1 MPa, to use as control.
77

USA). Cell dry weight (CDW) was routinely determined using a standard curve relating L. reuteri
optical density and CDW.

samples were prepared and aliquots of 1.0 mL of proper dilutions were plated in MRS agar (using
the pour plate technique), and incubated at 37 °C for 24 h.
5.2.6. Quantification of glycerol, glucose, and fermentation products

Culture samples were centrifuged at 8,000 g and 4 ºC for 10 min and the collected
using a HPLC Knauer system equipped with Knauer K-2301 RI and K-2501 UV detectors and an
Aminex HPX-87H cation exchange column (300 x 7.8 mm) (Bio-Rad Laboratories Pty Ltd,
Hercules, CA, USA). The mobile phase was 13 mM H 2SO4, delivered at a flow rate of 0.6 mL
min-1 and the column maintained at 65 °C. Peaks were identified by their retention times and
quantified using calibration curves prepared with different standards. Because of co-elution of
glycerol and lactate, and since glycerol is only detected by the RI detector, but lactate is detected
by both detectors, lactate concentration was estimated through the UV detector. The RI area
corresponding to the lactate concentration was estimated using the corresponding RI calibration
curve. Glycerol quantification was determined by the difference between the total area of the RI
peak and the calculated contribution of lactate in the area of the peak.
5.2.7. Analysis of the leakage of nucleic acids and proteins

Effect of HP on membrane permeability of L. reuteri was measured by determining the
contents of nucleic acids and extracellular proteins on extracellular medium, according to Dai et
al. (2017). After each fermentation cycle, changes of nucleic acid and extracellular protein
contents were determined based on their maximum absorption wavelength at 260 nm and 280 nm,
respectively. The absorbance (A) of the samples (fermented under HP conditions or at 0.1 MPa)
was measured at 260 nm and 280 nm after centrifugation at 8,000 g for 10 min. Nucleic acid or
extracellular protein contents (%) were calculated by comparison between the values obtained at
HP and the ones at 0.1 MPa, using the following equation:
78
𝐴ℎ𝑖𝑔ℎ 𝑝𝑟𝑒𝑠𝑠𝑢𝑟𝑒 − 𝐴0.1 𝑀𝑃𝑎

𝑁𝑢𝑐𝑙𝑒𝑖𝑐 𝑎𝑐𝑖𝑑 𝑜𝑟 𝑒𝑥𝑡𝑟𝑎𝑐𝑒𝑙𝑙𝑢𝑙𝑎𝑟 𝑝𝑟𝑜𝑡𝑒𝑖𝑛 (%) = × 100
𝐴0.1 𝑀𝑃𝑎
where Ahigh pressure corresponds to the absorbance of samples obtained at HP (10 or 25 MPa), and
A0.1 MPa corresponds to the absorbance of samples obtained at atmospheric pressure (0.1 MPa).

samples.
5.3. Results and discussion
5.3.1. Cell growth

To understand how different levels of sub-lethal HP affected L. reuteri cells throughout
the fermentation cycles, biomass concentration and viable cell counts were determined. Figure
5.2 shows the variation of both parameters after each fermentation cycle, at different pressure
conditions (10 and 25 MPa, as well as 0.1 MPa as control), for the samples with added acetate in
the culture medium (Figures 5.2.a and 5.2.b), and the ones without added acetate (Figures 5.2.c
and 5.2.d). In the course of these fermentation cycles, the initial acetate concentration and the
pressure applied were both found to affect cell growth.
In samples with initial acetate, biomass concentration showed a general decreasing trend
throughout the cycles at 0.1 and 10 MPa, indicating no improvement of cell growth. At 25 MPa,
biomass concentrations were always lower (p < 0.05) than at 0.1 and 10 MPa, and varied
differently over the cycles, with the highest concentration being reached at the second cycle (1.73
g L-1), and gradually decreasing thereafter. Still, biomass concentration at the fourth cycle (0.97
g L-1) was about 2-fold higher that the observed for first one (0.57 g L-1), which contrasts with the
trend at the other pressures, and suggests an improvement in cell ability to grow at 25 MPa.
Regardless of the pressure conditions applied, viable cell counts showed only slightly variation
throughout the cycles. In addition, cell counts were considerably higher at 10 and 25 MPa
compared to 0.1 MPa, which may result from differences in cell growth curves under different
pressure conditions.
In samples without added acetate, biomass concentrations at 0.1 MPa remained relatively
stable over the cycles, with a slight trend to increase (from 2.52 g L-1 in the first cycle, to 2.79 g
79
L-1 in the last one). Application of HP to the process, 10 or 25 MPa, led to lower (p < 0.05)
biomass concentrations in the first fermentation cycle: 0.66 and 0.24 g L -1 were observed for 10
and 25 MPa, respectively. However, biomass concentration was found to considerably increase
in the subsequent HP-cycles, indicating adaptation of L. reuteri cells to these conditions. At 10
MPa, biomass concentration increased to 2.87 g L-1 in the second cycle, and remained in this
range of values during the following ones. At 25 MPa, the highest biomass concentration (2.13 g
L-1) was reached at the third cycle, but decreased to 0.63 g L-1 in the fourth one, showing that the
eventual resistance and adaptation mechanisms developed at 25 MPa were not stable throughout
the fermentation cycles.
Figure 5.2. Biomass concentration and viable cell counts after each fermentation cycle (24 h), for samples
with an initial acetate concentration of 5 g L-1 (a and b), and for samples without initial acetate (c and d),
according to the pressure applied: 0.1 MPa (control samples), 10 MPa, and 25 MPa.
Similarly, the effects of HP on cell counts were more prominent in the first fermentation
cycle, with lower values at 10 and 25 MPa compared to 0.1 MPa. However, such as observed for
biomass concentration, cell growth was stimulated during the subsequent fermentation cycles,
reaching cell counts of 9.69 and 8.04 log10(colony forming units (CFU) mL-1) after the fourth
cycle at 10 and 25 MPa, respectively. These results suggest development of stress response
mechanisms to improve HP tolerance. Interestingly, this effect seems to be dependent of the initial
80
acetate concentration in the culture medium, since it was solely observed for samples without
acetate.
Pavlovic et al. (2008) have also reported adaptation of a lactobacilli strain to HP, by
isolating a piezotolerant Lactobacillus sanfranciscensis after incubation at 50 MPa during 25
cycles. Compared to the wild type, that strain showed enhanced ability to grow at 50 MPa. The
authors proposed that piezotolerance was promoted by increased expression of ssrA, leading to
higher amounts of tmRNA, a small RNA that provides a quality control function for translation
by recognizing stalled ribosomes and targeting the incomplete proteins for degradation.
Therefore, tmRNA might help to prevent accumulation of truncated, potentially harmful proteins
at higher pressures and make proteolysis more efficient (Pavlovic et al., 2008). Several other
genes were reported to be upregulated under sub-lethal HP conditions, in different
microorganisms. These include lipid synthesis genes, to increase the proportion of unsaturated
fatty acids in the cellular membrane and thus improve fluidity, and chaperone-encoding genes, to
help in maintaining protein folding and thus protein function (Oger and Jebbar, 2010). Increased
growth under HP may be also promoted by expression of HP-specific genes (Kato and Qureshi,
1999), and adaptation of the biomolecules structure to sustain pressure (Chilukuri and Bartlett,
1997). Most of these pressure-adaptation mechanisms would affect not only the cell ability to
grow under pressure, but also the overall metabolic and biological profile of the microorganism.
With this in mind, the main substrates and products involved in L. reuteri fermentation were
evaluated, being the results reported in further sections.
5.3.2. Substrate consumption

During fermentation, L. reuteri simultaneously consumes two carbon sources: glucose,
which is directed to the glycolysis pathway and results in the production of several products (e.g.
lactate, acetate and ethanol); and glycerol, which is metabolized by a different pathway, using
enzymes encoded in its propanediol-utilization (pdu) operon (Zaushitsyna et al., 2017). Therefore,
the consumption of both substrates was monitored throughout the fermentation cycles, at different
pressure conditions. The results are presented in Figure 5.3, for samples with added acetate in the
culture medium (Figures 5.3.a and 5.3.b), and for the ones without added acetate (Figures 5.3.c
and 5.3.d). In general, the uptake of glycerol and glucose showed similar profiles throughout the
cycles, as already observed in previous studies (Doleyres et al., 2005; Jolly et al., 2014; Lüthi-
Peng et al., 2002). This effect is explained by the dependency between them, since the NADH
generated from glucose by the glycolysis pathway is recycled by the presence of glycerol and its
transformation into 1,3-PDO (Doleyres et al., 2005).
At 0.1 MPa, the samples with acetate showed only slight variations in substrate uptake
over the cycles, for both glycerol and glucose. At 10 MPa, substrate uptake was lower (p < 0.05)
81
in the first cycle, but increased in the following ones, reaching residual concentrations similar (p
> 0.05) to 0.1 MPa at the fourth cycle. This increase in substrate consumption at 10 MPa was not
related to cell growth, since biomass concentration was reduced over the cycles at this pressure
(Figure 5.2.a). This indicates enhanced production of specific metabolites at 10 MPa, which may
be related to the L. reuteri response to pressure. At 25 MPa, substrate uptake seems to be inhibited,
being usually lower (p < 0.05) than at 0.1 and 10 MPa. The highest uptake was reached in the
second cycle, but it decreased thereafter in the subsequent cycles, similarly to the observed for
cell growth (Figure 5.2.a).
Figure 5.3. Glycerol and glucose concentrations after each fermentation cycle (24 h), for samples with an
initial acetate concentration of 5 g L-1 (a and b), and for samples without initial acetate (c and d), according
to the pressure applied: 0.1 MPa (control samples), 10 MPa, and 25 MPa.
Regarding the samples without initial acetate, substrate uptake increased over the
fermentation cycles at 0.1 MPa. At 10 MPa, substrate consumption was low in the first cycle, but
considerably increased in the subsequent ones. At the fourth cycle, the residual glycerol and
glucose concentrations were identical (p > 0.05) to the ones observed at the correspondent cycle
at 0.1 MPa. This pronounced increase in substrate consumption can be related to metabolic
changes in response to pressure, and even development of adaptation mechanisms to improve the
L. reuteri ability to grow and ferment under pressure. Fermentation cycles at 25 MPa showed
82
similar behavior, with low substrate uptake in the first cycle, and a subsequent increase in the
second and third cycles. However, glycerol and glucose consumption decreased in the fourth one,
with the residual concentrations coming closer to the ones in the first cycle. This effect of
“improvement and decline” was detected in both fermentation samples (with and without acetate)
at 25 MPa, but at different timelines: in samples with acetate, the peak of substrate consumption
was at the second cycle, while in samples without acetate the peak was at the third one. A similar
effect was also observed for Saccharomyces cerevisiae in another work of our research group
(unpublished results).
Therefore, application of consecutive fermentation cycles was found to improve substrate
consumption of both glycerol and glucose. This improvement was observed not only under HP
conditions, but also at 0.1 MPa. However, the enhancement of substrate consumption was more
pronounced under HP, particularly at 10 MPa. The samples without acetate were also more prone
to enhance substrate consumption and, thus, they seem more able to adapt to HP. The increase in
glucose consumption over the cycles might be associated to cell growth, but only when biomass
concentration was also increased, e.g. in samples without acetate. The cases with no cell growth
throughout the cycles (e.g. samples with acetate) imply that the surplus of glucose consumed has
been used to other purposes, such as increased production of metabolites involved in cell
maintenance, with functions that include osmoregulation, cell motility, turnover of
macromolecular compounds and defense mechanisms (Van Bodegom, 2007). Since glycerol is
not a carbon source for growth, the increase in glycerol uptake is certainly not related to cell
growth, and might indicate the increased production of some glycerol-derived compounds, such
as 1,3-PDO.
5.3.3. Production of 1,3-propanediol

In our previous study with L. reuteri, the production of 1,3-PDO was found to be
stimulated during fermentation at 10 MPa (one 32 h-cycle), in samples with acetate (unpublished
results). In the present work, we intended to assess if the application of consecutive cycles of
fermentation under pressure could improve even further the production of this bio-compound.
The results for these 24 h-fermentation cycles are presented in Figure 5.4, for samples with added
acetate in the culture medium (Figure 5.4.a), and for those without added acetate (Figure 5.4.b).
The 1,3-PDO yields (Y1,3-PDO/Gly, g g-1) and productivities (Q1,3-PDO, g L-1 h-1) were also estimated,
as indicated in Table 5.1.
When fermentation was performed at 0.1 MPa, the production of 1,3-PDO increased
between the first and the fourth cycles, regardless of the presence/absence of initial acetate. As
consequence, higher 1,3-PDO titers, yields and productivities were achieved at the fourth cycle,
compared to the first one. This positive effect at 0.1 MPa demonstrated that the use of consecutive
83
fermentation cycles itself stimulates the production of 1,3-PDO, as well as substrate consumption
and, in some cases, even cell growth. This can be explained by an adaptive process to the stresses
naturally involved in fermentation (even at 0.1 MPa), such as osmotic pressure, oxidative stress,
reduction of pH, and others (Serrazanetti et al., 2009).
Figure 5.4. 1,3-Propanediol concentration after each fermentation cycle (24 h), for samples with an initial
acetate concentration of 5 g L-1 (a), and for samples without initial acetate (b), according to the pressure
applied: 0.1 MPa (control samples), 10 MPa, and 25 MPa.
In fermentation cycles under HP, the profiles of samples with and without added acetate
were different from each other, and will be discussed separately. In the case of samples with
acetate, the production of 1,3-PDO at 10 MPa was usually higher (p < 0.05) than at 0.1 MPa. Over
the cycles, 1,3-PDO production gradually increased at both conditions, but more pronouncedly at
10 MPa. As consequence, the final titer at the fourth cycle was significantly higher (p < 0.05) at
10 MPa than at 0.1 MPa (4.59 and 4.16 g L-1, respectively). Yields and productivities were also
higher at 10 MPa, corresponding to improvements of ≈ 10 % relative to same cycle at 0.1 MPa.
It is worthy to note that, in our previous study, similar yield and productivity improvements were
already achieved at 10 MPa, for these type of samples (with acetate), after only one fermentation
cycle but with longer duration (32 h) (unpublished results). Therefore, although these
fermentation cycles under HP were favorable for the production of 1,3-PDO, it is arguable if this
corresponds to a meaningful improvement. It may not be worth to use it in this particular case,
since this approach involves a more complex, expensive and time-consuming processes.
However, adaptation to pressure might open new possibilities, with different metabolic features
and different process applications.
84
Table 5.1. 1,3-Propanediol yields on glycerol (Y1,3-PDO/Gly) and 1,3-PDO productivities (Q1,3-PDO), at the end
of each fermentation cycle (24 h), at different pressure conditions, in samples with and without acetate.
Samples Pressure Cycle Y1,3-PDO/Gly (g g-1) Q1,3-PDO (g L-1 h-1)
1 0.242 ± 0.002 0.123 ± 0.002

2C2 0.291 ± 0.006 0.155 ± 0.002
0.1 MPa
3 0.296 ± 0.002 0.161 ± 0.001
4 0.317 ± 0.003 0.173 ± 0.001
1 0.262 ± 0.003 0.128 ± 0.002
2C2 0.292 ± 0.018 0.156 ± 0.009
With acetate 10 MPa
3 0.325 ± 0.003 0.176 ± 0.002
24 0.350 ± 0.002 0.191 ± 0.001
13 0.041 ± 0.001 0.018 ± 0.001
24 0.306 ± 0.012 0.171 ± 0.005
25 MPa
3 0.245 ± 0.012 0.120 ± 0.004
4 0.244 ± 0.003 0.104 ± 0.001
1 0.290 ± 0.006 0.138 ± 0.001
2 0.347 ± 0.001 0.208 ± 0.001
0.1 MPa
3 0.327 ± 0.012 0.200 ± 0.006
4 0.346 ± 0.003 0.211 ± 0.002
1 0.016 ± 0.003 0.004 ± 0.001
2 0.272 ± 0.004 0.150 ± 0.005
Without acetate 10 MPa
3 0.356 ± 0.011 0.218 ± 0.007
4 0.397 ± 0.000 0.242 ± 0.006
1 0.016 ± 0.000 0.003 ± 0.002
2 0.145 ± 0.015 0.050 ± 0.010
25 MPa
3 0.236 ± 0.021 0.118 ± 0.007
4 0.135 ± 0.005 0.038 ± 0.002
Yields and productivities were calculated from a single time-point corresponding to the end of each cycle (24 h). Values reported in
the table represent the mean ± SD of two independent biological replicates, analyzed in duplicated.
At 25 MPa, the production of 1,3-PDO was considerably improved between the first and
second cycles (from 0.43 to 4.11 g L-1), but this effect was not stable and did not persist in the
following cycles. At the fourth cycle, a titer of 2.51 g L-1 was achieved, which is higher than the
obtained at the first cycle, but lower than the found at the second and third ones. At the moment,
there is still no explanation for this “improvement and decline” effect, but it suggests that L.
reuteri might not be able to adapt (at least not steadily) to a pressure of 25 MPa. In fact, continuous
85
application of this pressure may cause extensive damaging effects to the bacterial cells. For
instance, pressures between 10 – 26 MPa in E. coli are enough to impair relevant cellular
processes, such as motility, substrate transport and cell division (Oger and Jebbar, 2010).
If the use of consecutive fermentation cycles under HP did not seem to bring sufficient
gain in the samples with acetate, the same is not the case in the samples without acetate. At the
fourth cycle, fermentation at 10 MPa had significantly higher (p < 0.05) 1,3-PDO titer (5.80 g L-
1
), as well as higher yield and productivity (0.397 g g-1, and 0.242 g L-1 h-1 respectively), compared
to 0.1 MPa. These correspond to yield and productivity improvements of ≈ 15 %. In our previous
study with one 32 h-cycle (unpublished results), the samples without acetate at 10 MPa showed
higher 1,3-PDO yield (0.414 g g-1), but considerably lower titer (3.42 g L-1) and productivity
(0.107 g L-1 h-1) after 32 h. Therefore, in samples without acetate, the consecutive cycles of
fermentation at 10 MPa seem to have a positive effect on L. reuteri fermentation, in terms of 1,3-
PDO production. In contrast, over the cycles at 25 MPa, the 1,3-PDO production was always
lower than at 0.1 MPa (p < 0.05). The highest titer (2.82 g L-1) was achieved at the third cycle,
and then decreased in the subsequent one, thus indicating that L. reuteri was not able to adapt to
these pressure conditions, at least within the number of cycles studied.
The 1,3-PDO titers reached in this work are in the range of some of the values reported
in literature for L. reuteri glycerol/sugar co-fermentation (≈ 4 g L-1) (El-Ziney et al., 1998; Ragout
et al., 1996). However, these are considerably lower compared to other fermentation studies with
LAB. For instance, Lactobacillus diolivorans has been reported to produce 1,3-PDO
concentrations as high as 74 g L-1 or 92 g L-1 in fed-batch experiments (Lindlbauer et al., 2017;
Pflügl et al., 2012). For fermentation with L. reuteri, Jolly et al. (2014) reached 1,3-PDO
concentrations of 65.3 g L-1 during fed-batch experiments, while Baeza-Jiménez et al. (2011) and
Ricci et al. (2015) obtained concentrations of 29 and 41 g L-1, respectively, for batch cultivations.
Nevertheless, it is important to note that all of these studies were performed in bioreactors, with
monitoring, adjustment and optimization of several important process parameters. In the present
work, fermentation was carried out in batch, with different and simpler experimental set-up, as a
proof of concept to evaluate if it was possible to carry out the process under these pressure
conditions, and if it could promote any improvement.
In short, application of consecutive fermentation cycles at 10 MPa stimulated the
production of 1,3-PDO, regardless of the presence/absence of acetate in the initial culture
medium. However, the improvement of 1,3-PDO production was more pronounced in samples
without acetate, with an increase of 1,3-PDO titer, yield and productivity by 15 % at the fourth
fermentation cycle, relative to the same cycle at 0.1 MPa. By comparing the use of fermentation
cycles at 10 MPa with “conventional fermentation”, i.e. without the fermentation cycles and at
0.1 MPa, improvements of 52 %, 37 %, and 75 % were achieved for titer, yield and productivity,
86
respectively. This represents a meaningful process enhancement and can have a real impact in the
biotechnological production of 1,3-PDO. The reasons behind these improvements are not
completely understood, but might be related to the development of general and specific stress
responses, as well as adaptive mechanisms triggered by repeated exposure to sub-lethal HP.
5.3.4. Production of acetate, lactate and ethanol

Lactate, acetate, and ethanol are important reaction by-products obtained from glucose
metabolism. The effects of consecutive fermentation cycles (at 0.1, 10 and 25 MPa) on these bio-
chemicals are represented in Figure 5.5 – for lactate (Figures 5.5.a and 5.5.d), acetate (Figures
5.5.b and 5.5.e) and ethanol (Figures 5.5.c and 5.5.f). The yields (Y, g g-1) of each of these
compounds on glucose are indicated in Table 5.2.
In general, the production of all these by-products increased through the fermentation
cycles, but with specific profiles that varied according to the by-product in question. In samples
with acetate, by-product production at 10 MPa was generally similar to 0.1 MPa, with some slight
differences throughout the cycles: for instance, at the fourth fermentation cycle, acetate
concentration was higher (p < 0.05) at 10 MPa than at 0.1 MPa, while the opposite effect (p <
0.05) was observed for lactate and ethanol. The yields estimated for each by-product confirmed
this effect, and indicate a metabolic shift promoted by the consecutive fermentation cycles at 10
MPa, favoring the production of acetate and impairing the production of lactate and ethanol. In
samples without acetate, the by-product production at 10 MPa was low during the first two cycles
(similarly to the production of 1,3-PDO), but reached values similar to 0.1 MPa in the subsequent
cycles. In these last cycles, glucose metabolism have also shifted to the production of acetate,
such as observed in the samples with initial acetate. These metabolic changes at 10 MPa may be
related to the regulation of the NAD+/NADH ratio, since the formation of lactate and ethanol
competes with glycerol-derived 3-HPA for NADH. Consequently, if the cells are able to convert
3-HPA into 1,3- PDO, a lower amount of reduced fermentation products (lactate and ethanol) are
formed, and a higher amount of oxidized products (acetate) can be obtained (Ricci et al., 2015).
For that reason, the production of 1,3-PDO and acetate were positively correlated: the production
of 1,3-PDO at 10 MPa was improved over the cycles (relative to 0.1 MPa), and promoted the
formation of acetate, while reducing the formation of lactate and ethanol.
At 25 MPa the effect on by-product formation was highly variable according to the
presence or absence of acetate in the medium. In samples with acetate, lactate production was
significantly lower (p < 0.05) during all cycles, compared to 0.1 and 10 MPa.
87
Figure 5.5. Lactate, acetate and ethanol concentrations after each fermentation cycle (24 h), for samples
with an initial acetate concentration of 5 g L -1 (a, b and c), and for samples without initial acetate (d, e and
f), according to the pressure applied: 0.1 MPa (control samples), 10 MPa, and 25 MPa.
Acetate and ethanol production were also lower (p < 0.05) at the first cycle, but increased
in the following ones, and reached the range of concentrations obtained at 0.1 and 10 MPa. At the
fourth cycle, acetate concentration at 25 MPa was lower (p < 0.05) than for other pressures, but
ethanol concentration was significantly higher (p < 0.05). In fact, the highest ethanol
concentrations and yields were obtained for these samples at 25 MPa, at the third and fourth
cycles. Once again, the explanation might be related to the regulation of the NAD +/NADH ratio.
At 25 MPa, the production of 1,3-PDO was generally low (Figure 5.4), possibly due to inhibition
of one or more reactions involved in the glycerol-metabolizing pathway. If the conversion of 3-
HPA to glycerol is inhibited, NADH generated by glycolysis has to be drained to other pathways,
88
such as the acetylphosphate reduction to ethanol, which would be preferred over lactate, because
it allows re-oxidation of twice the amount of NADH. However, this higher ethanol production at
25 MPa was not observed in the samples without acetate. In fact, these samples accumulated
lower concentrations of all by-products, compared to 0.1 and 10 MPa (except ethanol, in the third
cycle).
Table 5.2. Yields of lactate (YLact/Glu), acetate (YAcet/Glu), and ethanol (YEtOH/Glu) on glucose, at the end of
each fermentation cycle (24 h), at different pressure conditions, in samples with and without acetate.
Samples Pressure Cycle YLact/Glu (g g-1) YAcet/Glu (g g-1) YEtOH/Glu (g g-1)
1 0.466 ± 0.010 0.152 ± 0.010 0.084 ± 0.000

2 0.536 ± 0.002 0.242 ± 0.009 0.093 ± 0.001
0.1 MPa
3 0.549 ± 0.011 0.255 ± 0.009 0.085 ± 0.000
4 0.571 ± 0.003 0.284 ± 0.007 0.081 ± 0.001
1 0.445 ± 0.013 0.146 ± 0.001 0.080 ± 0.001
2 0.521 ± 0.034 0.238 ± 0.030 0.088 ± 0.000
With acetate 10 MPa
3 0.568 ± 0.005 0.294 ± 0.007 0.096 ± 0.001
4 0.558 ± 0.002 0.315 ± 0.006 0.075 ± 0.001
1 0.134 ± 0.001 - 0.016 ± 0.000
2 0.468 ± 0.015 0.297 ± 0.010 0.116 ± 0.002
25 MPa
3 0.449 ± 0.023 0.286 ± 0.025 0.149 ± 0.001
4 0.463 ± 0.002 0.314 ± 0.003 0.129 ± 0.001
1 0.477 ± 0.012 0.212 ± 0.001 0.073 ± 0.002
2 0.506 ± 0.007 0.268 ± 0.001 0.058 ± 0.001
0.1 MPa
3 0.483 ± 0.013 0.251 ± 0.008 0.045 ± 0.001
4 0.475 ± 0.001 0.262 ± 0.006 0.042 ± 0.002
1 0.153 ± 0.014 0.135 ± 0.015 0.016 ± 0.002
2 0.335 ± 0.011 0.183 ± 0.006 0.038 ± 0.001
3 0.469 ± 0.006 0.279 ± 0.013 0.042 ± 0.001
4 0.456 ± 0.000 0.313 ± 0.006 0.034 ± 0.000
1 0.043 ± 0.032 0.087 ± 0.000 0.004 ± 0.000
2 0.260 ± 0.033 0.132 ± 0.002 0.037 ± 0.008
25 MPa
3 0.348 ± 0.041 0.197 ± 0.019 0.053 ± 0.002
4 0.283 ± 0.028 0.154 ± 0.010 0.032 ± 0.006
Yields were calculated from a single time-point corresponding to the end of each cycle (24 h). Values reported in the table represent
the mean ± SD of two independent biological replicates, analyzed in duplicated.
89
Overall, variations in by-product formation provided useful information about the L.

reuteri metabolism under HP. Fermentation at 0.1 and 10 MPa showed an increasing by-product
production over the cycles, which resulted in higher concentrations and yields. This stimulating
effect throughout the cycles was also observed for 1,3-PDO production, and, in some cases, for
cell growth. At 10 MPa, these improvements in product (and biomass) formation were also
accompanied by an increase in substrate uptake. However, at 0.1 and 25 MPa, changes in substrate
consumption were low, or even inexistent, which suggests that improved product formation in the
later cycles resulted from a more efficient energy use by L. reuteri cells. As for the specific effects
of HP, the use of consecutive cycles at these conditions promoted shifts in metabolic selectivity,
which differed according to the applied pressure level: at 10 MPa, the production of acetate tended
to increase over the cycles (such as previously observed for 1,3-PDO), while lactate and ethanol
tended to decrease; at 25 MPa, the formation of all by-products tended to decrease, with exception
of ethanol in the samples with acetate. To better understand how these fermentation cycles
affected metabolic selectivity, molar ratios between 1,3-PDO and by-products were estimated for
the four fermentation cycles performed at different pressure conditions (0.1, 10 and 25 MPa), in
the presence and absence of initial acetate (Table 5.3).
The ratios showed a general upward trend over the cycles at 0.1 and 10 MPa, but not at
25 MPa. In addition, the ratios at 10 MPa were always higher than at the other conditions, possibly
due to a shift in the metabolism towards the production of 1,3-PDO. High pressure may directly
affect the conversion of glycerol into 1,3-PDO, or may affect it indirectly, by modifying glucose
metabolism, which is closely related to glycerol metabolism. In fact, there are some reports in
literature about HP affecting the energy metabolism, particularly in the case of glycolysis (Bucka-
Kolendo and Sokołowska, 2017; Picard et al., 2007). Therefore, it is possible that some of the HP
effects observed in the present work result from changes in glycolysis and glucose metabolism.
The information regarding glycerol metabolism under HP is still scarce, and thus it is not certain
how pressure impacts this metabolic pathway. In order to understand it, further studies are of
interest, using, for instance, proteomic and metabolomics tools.
90
Table 5.3. Molar ratios between 1,3-PDO and by-products (lactate, acetate, and ethanol) produced after
each fermentation cycle (24 h), at different pressure conditions, in samples with and without acetate.
Molar ratio
Samples Pressure Cycle
1,3-PDO:by-products
1 0.242 ± 0.002
2 0.257 ± 0.001
0.1 MPa
3 0.263 ± 0.002
4 0.273 ± 0.003
1 0.274 ± 0.001
2 0.263 ± 0.002
With acetate 10 MPa
3 0.270 ± 0.001
4 0.297 ± 0.001
1 0.109 ± 0.005
2 0.279 ± 0.004
25 MPa
3 0.221 ± 0.002
4 0.213 ± 0.001
1 0.367 ± 0.002
2 0.472 ± 0.003
0.1 MPa
3 0.488 ± 0.003
4 0.510 ± 0.009
1 0.078 ± 0.009
2 0.507 ± 0.001
3 0.522 ± 0.000
4 0.560 ± 0.005
1 0.097 ± 0.012
2 0.313 ± 0.028
25 MPa
3 0.414 ± 0.011
4 0.335 ± 0.013
Values reported in the table represent the mean ± SD of two independent biological replicates, analyzed in duplicated.
5.3.5. Leakage of proteins and nucleic acids

Several effects of HP on microbial cells are related to modification of cell membrane
composition and fluidity (Mota et al., 2013). Lipid membranes are particularly pressure sensitive,
due to its high compressible potential. Membrane permeability can be evaluated by the changes
in the content of nucleic acids and proteins in the extracellular medium. The increase of these
compounds in the extracellular medium suggests its leakage, indicating higher permeability of
the cell membrane. Figure 5.6 shows the leakage of nucleic acids and proteins contents (%) in the
91
extracellular medium after each fermentation cycle at 10 and 25 MPa, relative to control samples
(0.1 MPa). Other studies used this method on microorganisms submitted to ultrasound or
pressurized CO2 and obtained positive values for protein and nucleic acid leakage, relative to the
control samples (without treatment) (Dai et al., 2017; Hashemi et al., 2018; Yao et al., 2014).
However, in the present work, almost all samples at 10 and 25 MPa showed negative leakage
values, which showed that HP did not increase leakage of cell constituents compared to 0.1 MPa
- on the contrary, HP was found to decrease it. The implications of these results are still not
completely understood, but suggest modification of cell membrane permeability under HP
conditions. With increasing pressure, lipid bilayers lose fluidity and became impermeable to water
and other molecules, with modification of protein-lipid interactions (Mota et al., 2013). Another
possibility, only applicable to protein leakage, could be related to denaturation and aggregation
of proteins caused by HP, but this phenomenon is rather unlikely at these low pressure levels,
since partial protein denaturation usually occurs at pressures of 100 MPa or above (Huang et al.,
2014).
Figure 5.6. Ratio of extracellular nucleic acid and protein contents (%) after each fermentation cycle (24
h) at 10 and 25 MPa, relative to 0.1 MPa, for samples with an initial acetate concentration of 5 g L -1 (a and
b), and for samples without initial acetate (c and d).
92
Considerable differences were observed between samples with and without acetate in the
initial medium. In samples with acetate, negative values were obtained for all cycles at 10 and 25
MPa, indicating that these pressures decreased leakage of nucleic acids and proteins relative to
0.1 MPa, possibly due to loss of membrane fluidity and permeability. However, throughout the
cycles, the relative leakage values (%) tended to come closer to zero, indicating that the HP effect
on cell membrane was attenuated, as a possible adaptation to pressure. Under HP environments,
several cell structures and components may be modified, due to fine tuning of gene expression,
that is expected to play an important role in low pressure environments (below 40 MPa) (Oger
and Jebbar, 2010). For instance, there is evidence of some microorganisms upregulating the
expression of lipid synthesis genes (Allen et al., 1999; Fernandes et al., 2004; Iwahashi et al.,
2005; Vezzi et al., 2005). As a result, cells may change the composition of lipid membrane under
HP, through the increase of unsaturated fatty acid content, in order to increase membrane fluidity
(Winter and Jeworrek, 2009). Thus, it is possible that L. reuteri begins to adapt to HP through the
cycles, by modifying cell membrane and making it more fluid and permeable.
In samples without acetate, the first cycle at 10 MPa had a positive value for nucleic acid
leakage (5.33 %), indicating higher leakage compared to 0.1 MPa. The release of nucleic acids
may be a result of membrane damage and, in some cases, it might correlate with microbial
inactivation (Lu et al., 2014; Pillet et al., 2016; Yao et al., 2014). Therefore, the use of HP at 10
MPa in these samples affected membrane integrity, and it may have caused bacterial inactivation.
In fact, low biomass concentration was observed at these conditions, and cell counts were slightly
lower compared to 0.1 MPa (Figure 5.2). However, in the following cycles at 10 MPa, the relative
leakage of nucleic acids was negative (i.e. lower at 10 MPa than at 0.1 MPa), with tendency to
come closer to zero. Regarding protein leakage at 10 MPa, the values were negative for all cycles
(including the first one, which contrasts with the nucleic acid leakage at this cycle).
In samples without acetate fermenting at 25 MPa, the nucleic acid leakage was positive
for all cycles, and more pronounced at the first cycle (11.18 %). In fact, biomass concentration
and cell counts (Figure 5.2) were lower at this pressure, and the fermentative activity was reduced.
Therefore, in these samples, fermentation at 25 MPa seems to cause membrane damage and, to
some extent, cell inactivation. In the case of protein leakage, a negative value was observed for
the first cycle, while the following ones all showed positive values. The results for these latter
cycles are in accordance with the leakage of nucleic acids, and support the possibility of L. reuteri
inactivation and cell destruction. However, the lower protein leakage in the first cycle remains to
be explained.
It is also interesting to note that HP only increased leakage of cell content (relatively to
atmospheric pressure) in samples without acetate, which suggests higher sensitivity of L. reuteri
to pressure in these samples, compared to the ones with acetate. This effect was even clearer at
93
25 MPa, with little or no adaptive effect being visible over the cycles. This effect was observed
along the study, and suggests that acetate may have a role on cell protection against HP, or may
be involved in the stress responses triggered by HP.
All in all, the results obtained for nucleic acid and protein leakage provided a first insight
into the HP effects on L. reteuri cell membrane, and how adaptation to these conditions may affect
membrane permeability. However, this was evaluated by an indirect measurement and thus, to
confirm these results, a more accurate and in-depth analysis should be performed to L. reteuri cell
membrane, in order to evaluate the effect of HP not only in terms of permeability, but also
regarding its composition.
5.4. Conclusions
Application of HP (10 and 25 MPa) during consecutive fermentation cycles was found to
affect L. reuteri cell growth and fermentative profile. In general, 1,3-PDO production tended to
increase over the fermentation cycles, especially at 10 MPa, which showed the highest 1,3-PDO
concentrations. The most suitable conditions for 1,3-PDO production were 10 MPa, in culture
medium without acetate. In that case, the highest titers, yields and productivities were achieved
at the fourth cycle: titers increased 15 % relative to the respective sample at 0.1 MPa, and even
more (52 %) relative to the “conventional approach”, i.e. fermentation without cycles, at 0.1 MPa.
The reasons behind these improvements are not completely understood, but might be related to
development general and specific stress responses, as well as adaptive mechanisms triggered by
repeated exposure to sub-lethal HP. In future work, it will be important to assess what is behind
these HP effects using proteomic and metabolic tools, and also to discern how HP-cycles impact
the composition and permeability of cell membrane. Our preliminary study regarding the effect
of HP-cycles on protein and nucleic acid leakage suggests a general decrease of cell membrane
permeability at 10 and 25 MPa. However, in some cases L. reuteri seems to gradually increase
membrane permeability over the cycles, as a possible adaptation to HP. All these modifications
in cell membrane will affect the exchange of substances (both substrates and products) throughout
the cell, with certain effect on cell metabolism.
Overall, this work provided relevant information regarding L. reuteri fermentation and its
adaptation to sub-lethal HP conditions (10 and 25 MPa). It also confirmed that HP may be useful
to develop new strategies for production of 1,3-PDO, a relevant food ingredient and bio-chemical
intermediate, with interest for several industries. One of the main advantages of HP in this context
is related to the possibility of applying it continuously without inactivation problems or heat
generation. Additionally, it is only necessary to provide energy to generate the pressure (and not
to maintain it), and so application of HP stress during the whole fermentation process has minimal
94
energetic costs. Therefore, this approach can be easily applied to different fermentation processes,
resulting in a real impact of HP technology in the food and biotechnological fields.
95
5.5. References
Allen, E.E., Facciotti, D., Bartlett, D.H., 1999. Monounsaturated but not polyunsaturated fatty
acids are required for growth of the deep-sea bacterium Photobacterium profundum SS9 at
high pressure and low temperature. Appl. Environ. Microbiol. 65, 1710–1720.
Biebl, H., Menzel, K., Zeng, A.-P., Deckwer, W.-D., 1999. Microbial production of 1,3-
propanediol. Appl. Microbiol. Biotechnol. 52, 289–297.
Bucka-Kolendo, J., Sokołowska, B., 2017. Lactic acid bacteria stress response to preservation
processes in the beverage and juice industry. Acta Biochim. Pol. 64, 459–464.
Chilukuri, L.N., Bartlett, D.H., 1997. Isolation and characterization of the gene encoding single-
stranded-DNA-binding protein (SSB) from four marine Shewanella strains that differ in
their temperature and pressure optima for growth. Microbiology 143, 1163–1174.
Dai, C., Xiong, F., He, R., Zhang, W., Ma, H., 2017. Effects of low-intensity ultrasound on the
growth, cell membrane permeability and ethanol tolerance of Saccharomyces cerevisiae.
Ultrason. Sonochem. 36, 191–197.
Dishisha, T., Pyo, S.-H., Hatti-Kaul, R., 2015. Bio-based 3-hydroxypropionic-and acrylic acid
production from biodiesel glycerol via integrated microbial and chemical catalysis. Microb.
Cell Fact. 14, 200.
96

(accessed 3.15.18).
913–916.
Fernandes, P.M.B., Domitrovic, T., Kao, C.M., Kurtenbach, E., 2004. Genomic expression
pattern in Saccharomyces cerevisiae cells in response to high hydrostatic pressure. Febs
Lett. 556, 153–160.
Hashemi, S.M.B., Khaneghah, A.M., Saraiva, J.A., Jambrak, A.R., Barba, F.J., Mota, M.J., 2018.
Effect of ultrasound on lactic acid production by Lactobacillus strains in date (Phoenix
dactylifera var. Kabkab) syrup. Appl. Microbiol. Biotechnol. 102, 2635–2644.
Iwahashi, H., Odani, M., Ishidou, E., Kitagawa, E., 2005. Adaptation of Saccharomyces
cerevisiae to high hydrostatic pressure causing growth inhibition. FEBS Lett. 579, 2847–
2852.
Kato, C., Qureshi, M.H., 1999. Pressure response in deep-sea piezophilic bacteria. J. Mol.
698.
Katrlík, J., Voštiar, I., Šefčovičová, J., Tkáč, J., Mastihuba, V., Valach, M., Štefuca, V.,
97
Gemeiner, P., 2007. A novel microbial biosensor based on cells of Gluconobacter oxydans
for the selective determination of 1,3-propanediol in the presence of glycerol and its
application to bioprocess monitoring. Anal. Bioanal. Chem. 388, 287–295.
34, 32–39.
Lu, H., Patil, S., Keener, K.M., Cullen, P.J., Bourke, P., 2014. Bacterial inactivation by high-
voltage atmospheric cold plasma: Influence of process parameters and effects on cell
leakage and DNA. J. Appl. Microbiol. 116, 784–794.
Oger, P.M., Jebbar, M., 2010. The many ways of coping with pressure. Res. Microbiol. 161, 799–
809.
Pavlovic, M., Hormann, S., Vogel, R.F., Ehrmann, M.A., 2008. Characterisation of a
piezotolerant mutant of Lactobacillus sanfranciscensis. Zeitschrift Fur Naturforsch. Sect. B
- a J. Chem. Sci. 63, 791–797.
Pillet, F., Formosa-Dague, C., Baaziz, H., Dague, E., Rols, M.-P., 2016. Cell wall as a target for
bacteria inactivation by pulsed electric fields. Sci. Rep. 6, 19778.
98
Saxena, R.K., Anand, P., Saran, S., Isar, J., 2009. Microbial production of 1,3-propanediol:
Recent developments and emerging opportunities. Biotechnol. Adv. 27, 895–913.
Schütz, H., Radler, F., 1984. Anaerobic reduction of glycerol to propanediol-1,3 by Lactobacillus
brevis and Lactobacillus buchneri. Syst. Appl. Microbiol. 5, 169–178.
Talarico, T.L., Dobrogosz, W.J., 1989. Chemical characterization of an antimicrobial substance
produced by Lactobacillus reuteri. Antimicrob. Agents Chemother. 33, 674–679.
Van Bodegom, P., 2007. Microbial maintenance: A critical review on its quantification. Microb.
Ecol. 53, 513–523.
Vezzi, A., Campanaro, S., D’Angelo, M., Simonato, F., Vitulo, N., Lauro, F.M., Cestaro, A.,
Malacrida, G., Simionati, B., Cannata, N., 2005. Life at depth: Photobacterium profundum
genome sequence and expression analysis. Science 307, 1459–1461.
Vollenweider, S., Lacroix, C., 2004. 3-Hydroxypropionaldehyde: Applications and perspectives
of biotechnological production. Appl. Microbiol. Biotechnol. 64, 16–27.
Winter, R., Jeworrek, C., 2009. Effect of pressure on membranes. Soft Matter 5, 3157–3173.
Yao, C., Li, X., Bi, W., Jiang, C., 2014. Relationship between membrane damage, leakage of
intracellular compounds, and inactivation of Escherichia coli treated by pressurized CO2. J.
Basic Microbiol. 54, 858–865.
Zaushitsyna, O., Dishisha, T., Hatti-Kaul, R., Mattiasson, B., 2017. Crosslinked, cryostructured
Lactobacillus reuteri monoliths for production of 3-hydroxypropionaldehyde, 3-
hydroxypropionic acid and 1, 3-propanediol from glycerol. J. Biotechnol. 241, 22–32.
99
100
CHAPTER VI
Comparative metabolomic profiling of

Lactobacillus reuteri under high pressure
fermentation cycles
CHAPTER VI. Comparative metabolomic profiling of L. reuteri under high pressure fermentation cycles
6.1. Introduction
Stress is one of the major driving forces of microbial evolution and adaptation
(Serrazanetti et al., 2009). When microorganisms are exposed to sub-lethal stress conditions, such
as high pressure (HP), general and specific stress response mechanisms are activated to allow
adaptation to the new conditions (Huang et al., 2014; Lado and Yousef, 2002). These stress
responses involve a complex and coordinated expression of genes, which will reflect upon the
microbial activity and, thus, on its metabolome (Serrazanetti et al., 2009). The metabolome is
formally defined as the collection of all small molecule metabolites or chemicals that can be found
in a cell, organ or organism. These small molecules can include a range of endogenous and
exogenous compounds, such as peptides, amino acids, nucleic acids, carbohydrates, organic acids,
vitamins, polyphenols, alkaloids, minerals and just about any other chemical that can be used,
ingested or synthesized by a given cell or organism (Wishart, 2008). The effects of stress on
microbial metabolome can be particularly relevant in the case of fermentative microorganisms
(i.e. microbial strains involved in fermentation processes), since these metabolic changes will
certainly affect the bioproducts and the bioprocess itself.
Metabolomics is an emerging field of “omics” research concerned with the high-
throughput identification and quantification of small molecule (< 1500 Da) metabolites in the
metabolome (German et al., 2005). The techniques most commonly used in metabolomics are:
mass spectrometry (MS) as an analytical technique, together with gas or liquid chromatography
(GC or LC, respectively) as additive methods; or nuclear magnetic resonance (NMR)
spectroscopy (Kruk et al., 2017). Each technique has their own advantages and disadvantages, as
indicated in Table 6.1. Whereas MS measures the ratio of mass to charge of ionized particles,
nuclear magnetic spectroscopy takes advantage of the magnetic properties of certain nuclei, such
as 1H, 13C, 31P and others (Brennan, 2014; Serkova and Niemann, 2006; Zhang et al., 2012). NMR
consists of exciting nuclear spins in an external magnetic field and then recording the
electromagnetic radiation emitted. When a given frequency of the electromagnetic wave is used
(dependent on the strength of the external magnetic field) only the nuclei with such resonance
frequency absorb it. As the local chemical environment of a nucleus affects its resonance
frequency it possible to distinguish nuclei in different chemical groups (the chemical shift) in a
given compound by NMR. In addition, under certain conditions, peak intensities can be used to
quantify the compounds present in the samples.
103
Table 6.1. Comparison between different metabolomics technologies (NMR spectroscopy, GC-mass
spectrometry, and LC-mass spectrometry). Adapted from Wishart (2008).
Technology Advantages Disadvantages
NMR spectroscopy  Quantitative;  Not very sensitive;

 Non-destructive;  Expensive instrumentation;
 Fast (2-3 min/sample);  Large instrument footprint;
 Requires no derivitization;  Cannot detect or identify salts
 Requires no separation; and inorganic ions;
 Detects all organic classes;  Cannot detect non-protonated
 Allows identification of novel compounds;
chemicals;  Requires larger (0.5 mL)
 Robust, mature technology; samples.
 Can be used for metabolite
imaging;
 Large body of software and
databases for metabolite
identification;
 Compatible with liquids and
solids.
GC-mass spectrometry  Robust, mature technology;  Sample not recoverable;

 Relatively inexpensive;  Requires sample derivitization;
 Quantitative (with calibration);  Requires separation;
 Modest sample size need;  Slow (20-30 min/sample);
 Good sensitivity;  Cannot be used in imaging;
 Large body of software and  Novel compound identification
databases for metabolite is difficult.
identification;
 Detects most organic and some
inorganic molecules;
 Excellent separation
reproducibility.
LC-mass spectrometry  Superb sensitivity;  Sample not recoverable;

 Very flexible technology;  Not very quantitative;
 Detects most organic and some  Expensive instrumentation;
inorganic molecules;  Slow (20-30 min/sample);
 Minimal sample size  Poor separation resolution and
requirement; reproducibility (vs. GC);
 Can be done without  Less robust instrumentation
separation (direct injection); than NMR or GC-MS;
 Has potential for detecting  Limited body of software and
largest portion of metabolome. databases for metabolite
identification;
 Novel compound identification
is difficult.
104
Over the past few years, two distinct approaches have emerged for processing and
interpreting metabolomics data. In one version (called the chemometric or non-targeted
approach), chemical compounds are not generally identified, and only their spectral patterns and
intensities are recorded, statistically compared and used to identify the relevant spectral features
that distinguish sample Classes (Nicholson et al., 1999; Trygg et al., 2007). These statistical
comparisons and feature identification techniques usually involve unsupervised clustering
(Principal Component Analysis or PCA) or supervised classification (Partial Least Squares
Discriminant Analysis or PLS-DA). While chemometric approaches like PCA and PLS-DA, on
their own, do not permit direct identification or quantification of compounds, they still allow an
unbiased (or untargeted), chemically comprehensive comparison to be made among different
samples (Wishart, 2008). In the other approach to metabolomics (called quantitative
metabolomics or targeted profiling), the focus is on attempting to identify and/or quantify as many
compounds in the sample as possible. Once the constituent compounds are identified and
quantified, the data are then statistically processed (using PCA or PLS-DA) to identify the most
important biomarkers or informative metabolic pathways (Weljie et al., 2006).
In the particular case of this work, we intended to analyze the general intracellular
metabolic changes in Lactobacillus reuteri DSM 20016 exposed to fermentation cycles under HP
(10 MPa). This may help to understand the effects of these HP-cycles on the L. reuteri
metabolome, and possibly disclose some mechanisms of pressure adaptation. Therefore, we
profiled the changes in the abundances of several metabolites (i.e. non-targeted strategy) using
1
H NMR spectroscopy. This technique was selected to perform this preliminary metabolomics
study, because it is fast, with no need for derivatization or preceding separation methods. In
addition, it detects all organic classes and allows identification of novel species, which could be
important in the course of the work.

Seed culture preparation and fermentation experiments were performed according to
Chapter V (section 5.2, Materials and methods). From these samples, specific case studies were
selected to perform metabolomics profiling, due to its pertinence to the work. These corresponded
to samples without acetate in the initial culture medium, collected after the first and fourth cycles
at 0.1 MPa (atmospheric pressure) and at 10 MPa. The samples without acetate were selected due
to the considerable adaptive effect observed throughout the cycles at 10 MPa, and to the higher
concentrations of 1,3-propanediol achieved.
105
6.2.2. 1H NMR experiments

Extracts were prepared by centrifugation (at 10,000 g for 15 minutes) followed by
filtration (0.45 m pore diameter) of the supernatants. Extracts were then dried in a vacuum
centrifuge, followed by storage in a desiccator until NMR analysis. Before NMR spectral
acquisition, samples were reconstituted using 600 L of phosphate buffer (100 mM, pH 3.0)
containing 0.01 % of 3-(trimethylsilyl)propionic-2,2,3,3-d4 acid, sodium salt (TSP-d4) as a
chemical shift and intensity reference. The mixture was then centrifuged (4500 g, 25 °C, 5 min)
and transferred into 5 mm NMR tubes to be analyzed.
1
H NMR spectra were recorded at 300 K on a Bruker Avance DRX 500 spectrometer
(Bruker BioSpin, Germany), operating at a proton frequency of 500.13 MHz, equipped with an
actively shielded gradient unit with a maximum gradient strength output of 53.5 G cm -1 and a 5
mm inverse probe. For each sample, a 1D 1H NMR spectrum was acquired using the noesypr1d
pulse sequence (Bruker pulse program library) with water presaturation. For all spectra, 128
transients were collected into 32,768 (32 K) data points with a spectral width of 10000 Hz, an
acquisition time of 3.3 s and relaxation delay of 5 s. Each free induction decay (FID) was zero-
filled to 64 k points and multiplied by a 0.3 Hz exponential line-broadening function prior to
Fourier transformation. iNMR software was used to manually phase and baseline correct the
spectra. The spectra were exported as a matrix using R-Studio in-house scripts and subsequently
normalised to TSP-d4. The spectra were overlaid and checked in iNMR to see whether alignment
was required. If required, the speaq package was used in R.
6.2.3. Multivariate data analysis

The multivariate analysis were applied to the aligned spectra, using the ropls package
(Thévenot et al., 2015) in R software (for statistical computing). Differences among samples were
visualized by Pareto-scaled for principal component analysis (PCA). The identification of
relevant metabolites was carried out by comparing the spectra with those of standard compounds
from the Biological Magnetic Resonance Data Bank and the Yeast Metabolome Database, and
confirmed by the Chenomx NMR Suite software. The relative amounts of the NMR metabolites
and the effect size were determined by integrating the area under the most well-separated
metabolite peak using in-house R scripts. Pairwise t-tests were carried out using the False
Discovery Rate (FDR) to adjust for multiple testing. Effect sizes were calculated and corrected
for small sample sizes using the formula:
3 𝑥1 − 𝑥2
𝐸𝑓𝑓𝑒𝑐𝑡 𝑠𝑖𝑧𝑒 = (1 − ( )) ( )
(4𝑛1 + 𝑛2 − 2) − 1 𝑝𝑜𝑜𝑙𝑒𝑑 𝑆𝐷
106
where pooled SD is the pooled standard deviation, x1 and x2 are the mean levels of metabolite x
and n1 and n2 are the number of replicates.

To obtain a wide perspective on how the HP fermentation cycles affected L. reuteri
metabolic profile, samples collected after the first and fourth cycles at 0.1 MPa (atmospheric
pressure) and at 10 MPa were analyzed by 1D 1H NMR and evaluated using different
metabolomic tools. It was intended not only to investigate the differences between fermentation
at 0.1 and 10 MPa, but also to analyze the evolution of each type of samples between the first and
fourth fermentation cycles. The characteristic 1D 1H NMR spectra of the samples obtained at
atmospheric pressure (C1 and C4, for cycles one and four, respectively) and at 10 MPa (P1 and
P4, for cycles one and four, respectively) are shown in Figure A.1 (Appendix A). Before peak
identification, a code name was assigned to each peak (M1 and M73) ranging from the higher to
lower ppm values. It is important to highlight that each metabolite can generate more than one
signal, when the compound has two or more protons in different “environments”. Spectral
comparisons with databases and an appropriate software were performed in an attempt to identify
some of the metabolites present in the samples. The obtained results are indicated in Table 6.2.
Table 6.2. List of relevant compounds identified in the samples by comparison with databases and an
appropriate software, with the respective chemical shifts and code names (attributed between M1 and
M73, from higher to lower ppm values).
Compounds Chemical shifts (ppm) Code name
1,3-Propanediol 3.66 – 3.73, 1.75 – 1.83 M44, M60

2,3-Butanediol 1.11 – 1.16 M69
Acetate 2.00 – 2.07 M58
Citrate 2.76 – 2.87, 2.66 – 2.76 M52, M53

Ethanol 1.26 – 1.30, 1.19 – 1.24 M67, M68
5.22 – 5.25, 4.53 – 4.70, 3.87 – 3.93, 3.73 – 3.75, M32, M36, M41, M43,
Glucose/glycerol
3.62 – 3.66, 3.53 – 3.59 M45, M46
Isobutyrate 2.36 – 2.43, 1.02 – 1.06 M56, M71
Lactate 4.12 – 4.22, 2.07 – 2.15, 1.30 – 1.38 M39, M57, M66
107
Insets for aromatic (6.0 - 9.0 ppm) and aliphatic (0.6 - 3.0 ppm) regions of the 1H NMR
spectra are presented (Figures A.1.b and A.1.c, respectively). The aromatic region is characteristic
of aromatic amino acids (such as phenylalanine, histidine, tyrosine and tryptophan), while the
aliphatic region is typically assigned to organic acids, alcohols, and aliphatic amino acids.
Therefore, peaks of L. reuteri specific fermentation products can be easily identified in the
aliphatic region, as indicated in Table 6.2. The region between 3.5 - 5.3 ppm is characteristic of
sugars (glucose, in this case) and glycerol, which are overlapped with other signals. For instance,
in this region, signals of 1,3-PDO (3.66 - 3.73 ppm) and lactate (4.12 - 4.22 ppm) were also
identified. By comparing the full spectra (Figure A.1.a) of all samples, slight differences can be
observed according to the pressure and the fermentation cycle. The most evident differences were
detected in the P1 samples, correspondent to the first fermentation cycle at 10 MPa. In fact, the
results of our previous work (Chapter V) showed that L. reuteri fermentative activity was very
low during the first fermentation cycle at 10 MPa.
In order to screen the differences of metabolites between fermentation samples, a
Principal Component Analysis (PCA) was performed from the 1D 1H NMR data. PCA is a
dimensional reduction technique that allows to easily plot, visualize and cluster multiple
metabolomic data sets based on linear combinations (known as principal axes) of their shared
features. As a clustering technique, PCA is most commonly used to identify how one sample is
different from another and which variables contribute most to this difference (Wishart, 2008). In
the present work, the PCA model showed a good fit of R2X (0.95). The score plot resulting from
PCA achieved by combining PC1 (81 % explained variance) and PC2 (14 % explained variance)
is shown in Figure 6.1, while the respective loadings are presented in Figure A.2 (Appendix A).
The PCA score plot revealed a clear separation between the first and the fourth cycles, by
primarily PC1. The loading plot (Figure A.2.a) revealed that glycerol and glucose seem to be the
major metabolites that positively contributed to PC1, while organic acids and alcohols (in
particular 1,3-PDO) negatively contributed to PC1. Therefore, the samples obtained after only
one fermentation cycle showed higher PC1, due to higher content of substrates (glycerol and
glucose) and lower content of typical fermentation products. In contrast, the samples obtained
after four fermentation cycles showed lower PC1, indicating higher content of fermentation
products and lower content of substrates, a profile characteristic of higher fermentative activity.
This suggests that the application of fermentation cycles (under HP or not) stimulated L. reuteri
fermentation, but with a considerably more pronounced effect at 10 MPa. The variables associated
to PC2 had less contribution (14 %) to discriminate the samples, but clearly separated the one-
cycle fermentation at 0.1 MPa from the remaining ones. The loading plot (Figure A.2.b) showed
that the main metabolites that contributed to PC2 were in the sugars and glycerol region (4.70 –
4.90 ppm; 3.50 – 4.00 ppm), as well as in the aliphatic region (1.00 – 2.00 ppm). However, these
108
metabolites were not clearly identified, and thus it was not possible to confirm the differences in
the metabolic profile of these samples.
Figure 6.1. PCA scores plot of L. reuteri fermentation metabolites, obtained by 1D 1H NMR. C1 and C4
samples correspond to the first and fourth fermentation cycles at atmospheric pressure, respectively; P1 and
P4 samples correspond to the first and fourth fermentation cycles at 10 MPa, respectively. In all cases, the
results are presented in duplicated (indicated as 1 and 2).
Changes in the metabolites levels in L. reuteri fermentation samples were visualized

according to a heat map (Figure 6.2). As expected, the heat map showed considerable differences
in the metabolic profile of the samples at 10 MPa after one fermentation cycle. In this case, higher
abundance of glucose and glycerol (M32, M45, M46) was observed, as well as other unidentified
metabolites (M24, M27, M47, M48). On the other hand, the metabolites with lower abundance in
these samples were 1,3-PDO and lactate, as well as M1 and M2 (unidentified).
109
Figure 6.2. Heat map of L. reuteri metabolites at different fermentation cycles and pressure conditions.
Scale is based on colors from red to blue representing an increase and a decrease in metabolite levels,
respectively. The C1 and C4 samples correspond to the first and fourth fermentation cycles at atmospheric
pressure, respectively; P1 and P4 samples correspond to the first and fourth fermentation cycles at 10 MPa,
respectively. The signals are indicated by a code name, between M1 and M73, from higher to lower ppm
values.
110
The profile of the one-cycle samples at 0.1 MPa showed less metabolite variation, with
only higher abundance of substrates (M45 and M46) and lower abundance of M24 and M27
(unidentified). In samples collected after the fourth fermentation cycle, the metabolic profiles at
0.1 and 10 MPa were clearly similar, such as already suggested by the PCA score plot. The main
differences may be attributed to M23, M24, M27 and M28, which are all in lower levels at 0.1
MPa. These metabolites were not identified, but seem to be part of the HP response and
adaptation, since showed lower abundance in both samples at atmospheric pressure.
To better understand the effects of fermentation cycles on L. reuteri metabolism, the
differential abundances between the first and the fourth cycles were analyzed using Volcano plots.
These plots permitted the visualization of the relationship between effect size and p-values. Figure
6.3 shows the differential abundances between the cycles, for fermentation at 0.1 MPa (Figure
6.3.a) and at 10 MPa (Figure 6.3.b).
In both cases, the fermentation cycles affected the metabolite abundance in the samples,
which suggested adaptation of L. reuteri to the fermentative conditions. At atmospheric pressure,
this effect may result from an adaptive process to the stresses naturally involved in fermentation,
such as osmotic pressure, oxidative stress, reduction of pH, and others (Serrazanetti et al., 2009).
The changes in metabolite abundance between the cycles at 0.1 MPa were subtle, but were
significantly more accentuated at 10 MPa. At this HP level, the metabolic profiles at the first and
at the fourth fermentation cycles are unequivocally different, which suggest adaptation to the
pressure conditions, possibly due to upregulation of general and specific sets of stress-response
genes (Oger and Jebbar, 2010), leading to a different metabolic profile. In fact, some of the
metabolites with higher relative abundance after the fourth cycle at 10 MPa (M1, M2, M16, M18,
M21, M55, M56, M63 and M71) were not assigned to any of the typical L. reuteri fermentative
substrates and products, and may be related to stress response mechanisms. From these, M16 and
M18 were correspondent to aromatic amino acids, probably phenylalanine, tyrosine, tryptophan.
In addition, M56 and M71 were assigned to isobutyrate, while the remaining signals were still not
assigned. In further studies, it will be relevant to properly identify these metabolic products, by
employing metabolomics tools suitable for that purpose.
In a different perspective, differential abundances between the samples at 10 and 0.1 MPa
(at the fourth cycle) were plotted in a Volcano plot (Figure 6.4). Similarly to the PCA score plot
(Figure 6.1) and to the heat map (Figure 6.2) presented above, the Volcano plot suggested similar
metabolic profiles at both pressure conditions, after four fermentation cycles. Therefore, these
fermentation cycles under pressure promoted adaptation of L. reuteri, resulting in a profile
comparable to the obtained at atmospheric pressure. Despite the similarities between the samples,
it was possible to identify two differentiating metabolites, each one specific for a pressure
condition. At 0.1 MPa (orientation to the right in the Volcano plot), the metabolite with slightly
111
Figure 6.3. Volcano plots showing the differences in the metabolic profiles of L. reuteri between the first
and the fourth fermentation cycles, at 0.1 MPa (a) and at 10 MPa (b). The signals/metabolites are indicated
by a code name, between M1 and M73, as previously indicated. The x-axis represents the effect sizes
(plotted on a log 2 scale) of the relative abundance of each metabolite between the samples after the first
and the fourth cycles. The y-axis represents the statistical significance p-value of the ratio fold-change for
each metabolite. Metabolites whose abundance is unchanged between the two samples will plot at the x-
axis origin. Metabolites that hyper-accumulate in one of the two samples under analysis will plot either to
the left (first cycle) or right (fourth cycle) of the x-axis origin.
112
Figure 6.4. Volcano plots showing the differences between the metabolic profiles of L. reuteri at 10 and
0.1 MPa, after the fourth fermentation cycle. The signals/metabolites are indicated by a code name, between
M1 and M73, as previously indicated.. The x-axis represents the effect sizes (plotted on a log 2 scale) of
the relative abundance of each metabolite between the samples after the first and the fourth cycles. The y-
axis represents the statistical significance p-value of the ratio fold-change for each metabolite. Metabolites
whose abundance is unchanged between the two samples will plot at the x-axis origin. Metabolites that
hyper-accumulate in one of the two samples under analysis will plot either to the left (10 MPa) or right (0.1
MPa) of the x-axis origin.
higher relative abundance is the M55 (unidentified), with chemical shift in the aliphatic region,
and thus it may correspond to an organic acid, alcohol, or aliphatic amino acid. On the other hand,
2,3-butanediol (2,3-BDO), indicated as M69, showed evident higher relative abundance at 10
MPa (orientation to the left in the Volcano plot).
2,3-Butanediol is a bulk chemical with multiple practical applications, such as the
production of synthetic rubber, plasticizers, fumigants, and also as an antifreeze agent, fuel
additive, octane booster, and many others. The production of bio-based 2,3-BDO is mainly
attained by Klebsiella pneumoniae, Klebsiella oxytoca and Bacillus polymyxa. It is produced from
pyruvate via several intermediate compounds, including α-acetolactate and acetoin (Celińska and
Grajek, 2009). A schematic representation of this metabolic pathway is shown in Figure 6.5.
113
Figure 6.5. Schematic representation of general metabolic pathways for 2,3-butanediol (2,3-BDO)
production by glucose metabolism, in lactic acid bacteria. Adapted from Gaspar et al. (2011) and Paul
(2010).
First, pyruvate from glycolysis can be either converted either into lactate in a reaction that
requires NADH (catalysed by lactate dehydrogenase, LDH) or into α-acetolactate (catalysed by
α-acetolactate synthase). α-Acetolactate is mostly produced under low NADH availability.
Further, α-acetolactate can be converted to acetoin by α-acetolactate decarboxylase, under
anaerobic conditions. Finally, butanediol dehydrogenase reduces acetoin to 2,3-BDO.
The production of 2,3-BDO by L. reuteri corresponds to an interesting metabolic feature,
which is poorly documented in literature, since the production of this compound is usually
attributed to other Lactobacillus strains - such as Lactobacillus brevis, Lactobacillus casei,
Lactobacillus helveticus, Lactobacillus plantarum. Because of that, the presence of 2,3-BDO was
not assessed in our previous studies with L. reuteri under HP (Chapters IV and V), but should
certainly be addressed in subsequent studies with this microorganism.
114
Another relevant aspect of this work concerns the formation of 1,3-PDO under the
different pressure conditions. In our previous study (Chapter V), 1,3-PDO production was
stimulated throughout the fermentation cycles at 10 MPa and, in consequence, the final titer at
the fourth cycle was significantly higher compared to 0.1 MPa. However, this effect was not
confirmed by the 1H NMR results. It is important to take into account that one of the main
disadvantages of 1H NMR corresponds to its low sensitivity, such as indicated in Table 6.1. In
this particular case, 1H NMR was used as a qualitative method, to perform general chemical
characterization and to provide a wide perspective on the metabolic profile of the samples. The
methodology applied was not the most suitable for quantitative analysis, in comparison to the
chromatographic methods (HPLC-RI/UV) applied in Chapter V, which used appropriate
standards and a suitable quantification methodology.
To further analyze this subject, metabolite plots showing the abundance of 1,3-PDO in
all samples are represented in Figure A.3 (Appendix A). The plots showed higher mean of 1,3-
PDO abundance at the fourth cycle at 10 MPa. However, due to the high standard deviation
values, these differences did not represent significant improvements relatively to the respective
cycle at 0.1 MPa. Therefore, it is highly probable that 1,3-PDO production was, in fact, stimulated
by the cycles at 10 MPa, such as suggested in our previous work (Chapter V), even if this was not
certainly confirmed by the 1H NMR results. The high standard deviation values can be explained
by the use of only two biological replicates for each sample, which were not enough to provide
more accurate results. Therefore, additional biological replicates should be included in future
metabolic profiling studies, in order to increase the accuracy of the analysis. In addition to that,
further studies should comprise 2D NMR experiments and comparison with appropriate
databases, to perform all peak assignments. With that information, it will be possible to fully
understand the impact of HP on L. reuteri metabolic profile, and to identify some of the pressure
adaptation mechanisms developed throughout the fermentation cycles.
6.4. Conclusions
The comparative metabolomic study between L. reuteri fermentation samples showed
distinct metabolic profiles according to the pressure applied (0.1 or 10 MPa) and the application
of fermentation cycles (one or four cycles). The PCA score plot revealed a clear separation
between the first and the fourth cycles, with the samples obtained after only one fermentation
cycle having higher abundance of fermentation substrates (glycerol and glucose), and the samples
obtained after four fermentation cycles having higher content of fermentation products (such as
organic acids and alcohols). This suggests that application of fermentation cycles (under HP or
not) stimulated L. reuteri fermentation, possibly due to development of adaptive mechanisms.
115
Visualization by Volcano plots showed that this adaptive effect throughout the cycles was
considerably more accentuated at 10 MPa. After four fermentation cycles, the metabolic profile
at 10 MPa was similar to 0.1 MPa, indicating that L. reuteri was able to adapt to pressure, with a
response comparable to the one observed at atmospheric pressure. Interestingly, one of the
metabolites characteristic of fermentation cycles at 10 MPa was 2,3-BDO, which showed higher
relative abundance at this pressure, relatively to the respective cycle at 0.1 MPa. Two mechanisms
were proposed to explain the stimulation of 2,3-BDO under HP: modification of lactose
dehydrogenase activity; and/or modification of the NAD +/NADH equilibrium.
Overall, the results of the present study unveil relevant information regarding the
adaptation of LAB to sub-lethal HP. These mechanisms are intrinsically related to the
development of different metabolic features, which certainly have an impact on the fermentation
process itself, and also on the resulting bioproducts. Therefore, HP can be applied to regulate
microbial metabolism to different purposes, through the optimization of pressure conditions able
to stimulate preferred metabolic pathways and/or to inhibit undesirable ones.
116
6.5. References
Brennan, L., 2014. NMR-based metabolomics: From sample preparation to applications in
nutrition research. Prog. Nucl. Magn. Reson. Spectrosc. 83, 42–49.
Celińska, E., Grajek, W., 2009. Biotechnological production of 2,3-butanediol - Current state and
prospects. Biotechnol. Adv. 27, 715–725.
Gaspar, P., Neves, A.R., Gasson, M.J., Shearman, C.A., Santos, H., 2011. High yields of 2,3-
butanediol and mannitol in Lactococcus lactis through engineering of NAD+ cofactor
recycling. Appl. Environ. Microbiol. 77, 6826–6835.
German, J.B., Hammock, B.D., Watkins, S.M., 2005. Metabolomics: Building on a century of
biochemistry to guide human health. Metabolomics 1, 3–9.
Kouassi, G.K., Anantheswaran, R.C., Knabel, S.J., Floros, J.D., 2007. Effect of high-pressure
processing on activity and structure of alkaline phosphatase and lactate dehydrogenase in
buffer and milk. J. Agric. Food Chem. 55, 9520–9529.
Kruk, J., Doskocz, M., Jodłowska, E., Zacharzewska, A., Łakomiec, J., Czaja, K., Kujawski, J.,
2017. NMR techniques in metabolomic studies: A quick overview on examples of
utilization. Appl. Magn. Reson. 48, 1–21.
Lado, B.H., Yousef, A.E., 2002. Alternative food-preservation technologies: Efficacy and
mechanisms. Microbes Infect. 4, 433–440.
Nicholson, J.K., Lindon, J.C., Holmes, E., 1999. “Metabonomics”: Understanding the metabolic
responses of living systems to pathophysiological stimuli via multivariate statistical analysis
of biological NMR spectroscopic data. Xenobiotica 29, 1181–1189.
Oey, I., 2016. Effects of high pressure on enzymes, in: Balasubramaniam, V., Barbosa-Cánovas,
G., Lelieveld, H. (Eds.), High Pressure Processing of Food. Food Engineering Series.
Springer, New York, US, pp. 391–431.
Oger, P.M., Jebbar, M., 2010. The many ways of coping with pressure. Res. Microbiol. 161, 799–
809.
Paul, B.J., 2010. Enhanced pyruvate to 2,3-butanediol production in lactic acid bacteria. US
2010/0112655 A1.
Serkova, N.J., Niemann, C.U., 2006. Pattern recognition and biomarker validation using
quantitative 1H-NMR-based metabolomics. Expert Rev. Mol. Diagn. 6, 717–731.
117

Thévenot, E.A., Roux, A., Xu, Y., Ezan, E., Junot, C., 2015. Analysis of the human adult urinary
metabolome variations with age, body mass index, and gender by implementing a
comprehensive workflow for univariate and OPLS statistical analyses. J. Proteome Res. 14,
3322–3335.
Trygg, J., Holmes, E., Lundstedt, T., 2007. Chemometrics in Metabonomics. J. Proteome Res. 6,
469–479.
Weljie, A.M., Newton, J., Mercier, P., Carlson, E., Slupsky, C.M., 2006. Targeted profiling:
Quantitative analysis of 1H NMR metabolomics data. Anal. Chem. 78, 4430–4442.
Wishart, D.S., 2008. Metabolomics: Applications to food science and nutrition research. Trends
Food Sci. Technol. 19, 482–493.
Zhang, A., Sun, H., Wang, P., Han, Y., Wang, X., 2012. Modern analytical techniques in
metabolomics analysis. Analyst 137, 293–300.
118
CHAPTER VII
The use of different fermentative approaches

on Paracoccus denitrificans: Effect of high
pressure and air availability on growth and
metabolism
Adapted from:
Mota, M.J., Lopes, R.P., Pinto, C.A, Sousa, S., Gomes, A.M., Delgadillo, I., Saraiva, J.A., The
use of different fermentative approaches on Paracoccus denitrificans: Effect of high pressure and
air availability on growth and metabolism. Submitted in Biotechnology Progress.
CHAPTER VII. The use of different fermentative approaches on P. denitrificans
7.1. Introduction
Biodiesel is a fuel originated from biomass, produced from vegetable oils and animal fats,
and represents one of the most promising alternatives to fossil fuels (da Silva et al., 2009). The
increasing biodiesel production is raising constrains that may compromise the sustainability of
the process, and one of the major problems corresponds to the formation of crude glycerol as a
reaction by-product (da Silva et al., 2009; Kolesárová et al., 2011). In order to maintain the
viability of biofuel economy, it became necessary to develop new and sustainable applications
for glycerol, such as the use as substrate for microbial fermentation processes, resulting in the
production of different value-added products, such as organic acids, alcohols, polymers, among
others (Mattam et al., 2013).
Paracoccus denitrificans is a Gram-negative microorganism able to grow in glycerol, as
well as in many other carbon sources, including methanol, ethanol, 1-butanol, and 1-pentanol
(Ueda et al., 1992; Yamane et al., 1996a, 1996b). The first strain of P. denitrificans was isolated
from soil more than one century ago by Beijerinck and Minkman (1910). It exhibits metabolic
versatility, and it was shown to grow both aerobically and anaerobically, performing complete or
partial denitrification. Air availability is a critical parameter in P. denitrificans growth and
fermentation. It affects not only cell growth, but also some other relevant metabolic features.
Kalaiyezhini and Ramachandran (2015) observed that P. denitrificans specific growth rates
increased with the increase in oxygen transfer rate, while moderate oxygen transfer rate promoted
poly(3-hydroxybutyrate) production. In aerobic bioprocesses, oxygen is a key substrate, and must
be continuously supplied (Garcia-Ochoa and Gomez, 2009). However, this dependence on
oxygen availability may be a limitation to some fermentation processes, particularly for high-
scale industrial processes. The requirement for high oxygen availability also presents a limitation
for the performance of fermentation under high pressure conditions since, currently, many high
pressure equipments are not adapted to allow continuous air supply. Therefore, in some specific
cases, it might be necessary to perform aerobic microbial processes under limited-air conditions.
The interest in exposing microbial cells to high pressure (HP) is related to growth
stimulation and/or improvement of fermentation (Mota et al., 2018). This approach involves the
use of sub-lethal HP levels that affect cell growth and metabolism, but without compromising cell
viability. In some cases, these modifications can represent considerable improvements, such as
increased yields, productivities and fermentation rates, lower accumulation of by-products and/or
production of different compounds. For instance, Picard et al. (2007) accelerated alcoholic
fermentation and increased ethanol yields in Saccharomyces cerevisiae by performing
fermentation at 5 and 10 MPa. High pressure was also tested to change the metabolic selectivity
of fermentative strains: application of pressures of 7 and 17 MPa during fermentation by
Clostridium thermocellum redirected the metabolism from the production of by-products (such as
121
acetate) to ethanol, compared to fermentation at atmospheric pressure (Bothun et al., 2004).

Another example is the use of HP to modify the properties of biopolymers produced during
fermentation. Production of bacterial cellulose by Gluconacetobacter xylinus under HP (30, 60
and 100 MPa) showed profound differences in morphological properties of the polymer
depending on the applied pressure conditions. The cellulose produced under HP had a
significantly higher density compared with the cellulose produced at atmospheric pressure (Kato
et al., 2007). Regarding polymer production, Follonier et al. (2012) applied a low pressure level
(0.7 MPa) to Pseudomonas putida KT2440 and enhanced productivity of medium-chain-length
polyhydroxyalkanoate production, even with a significant decrease in specific growth rates. The
effects of HP have also been evaluated in the context of food fermentation: on lactic acid
fermentation, for production of probiotic yogurt (Mota et al., 2015); and in the beginning of
malolactic fermentation by Oenococcus oeni (Neto et al., 2016). In the first case, HP was found
to reduce the fermentation rate, but it was still possible to produce yogurt under pressure by
extension of the fermentation time (Mota et al., 2015). The probiotic yogurt produced at 5 MPa
showed different biochemical composition (unpublished results), and possibly different
organoleptic properties. In the study with O. oeni (microorganism used by the wine industry to
perform malolactic fermentation), the strain was able to perform fermentation during and after
HP-stresses of 50 and 100 MPa, with some metabolic changes. For instance, the HP-stress of 100
MPa stimulated the production of the D-lactate isomer, relative to the L-isomer.
There is still a great potential to explore in this field, with the studies conducted so far
showing promising results, not only regarding food fermentations, but also for biotechnological
processes. Considering that HP can be a useful tool for improving the glycerol-based fermentation
processes, the present work was divided in two main goals: i) perform a preliminary study
regarding the effects of air availability on P. denitrificans growth and metabolism; ii) assess if P.
denitrificans is able to grow and maintain metabolic activity under HP (10, 25 and 35 MPa), even
with limitations in terms of volume and air supply.

A lyophilized culture of Paracoccus denitrificans DSM 413 (ATCC 17741), obtained
from Deutsche Sammlung von Mikroorganismen und Zellkulturen (DSMZ, Braunschweig,
Germany), was used in this study. The strain was reconstituted on nutrient broth according to the
manufacturer’s instructions. The strain was sub-cultured on nutrient agar plates and incubated at
30 ºC for 24 h and then preserved at 4 ºC for a maximum period of 1 month.
122
Rich Medium reported by Hori et al. (1994) was used for inoculum preparation. It
included polypeptone (10 g L-1), yeast extract (10 g L-1), meat extract (5 g L-1) and (NH4)2SO4 (5
g L-1). Mineral medium reported by Mothes et al. (2007) was used for the fermentation
experiments. It contained glycerol (20 g L-1), yeast extract (4.5 g L-1), K2HPO4 (5 g L-1), KH2PO4
(0.5 g L-1), CaCl2•2H2O (20 mg L-1), MgSO4•7H2O (1 g L-1), and trace elements solution (2 mL
L-1). The composition of trace elements solution is as follows: FeSO 4•7H2O (4.98 g L-1), ZnCl2
(0.44 g L-1), CuSO4•5H2O (0.78 g L-1), Na2MoO4•2H2O (0.24 g L-1), MnSO4•4H2O (0.81 g L-1),
dissolved in 1 N HCl solution.
7.2.2. Inoculum preparation and inoculation

A single cell colony was seeded into 100 mL of rich medium and incubated overnight at
35 ºC for 16 – 20 h, in a rotary incubator (160 rpm). Mineral medium was inoculated with 5 %
(v/v) of standard inoculum, in an aseptic environment, within a laminar flow cabinet (BioSafety
Cabinet Telstar Bio II Advance, Terrassa, Spain), to avoid contamination.
7.2.3. Fermentation experiments: Effect of air availability

For these experiments, three different types of samples were prepared: i) samples “with
air”, which performed fermentation in shake-flasks, with medium:air volume ratio (Vmedium:Vair
ratio) of 1:5, and agitation speed of 135 rpm; ii) samples “without air”, which fermented in
polyethylene bags, sealed with no air; iii) samples “24 h with air + 48 h without air”, which
fermented in shake flasks (with air availability) during the first 24 h, and then were transferred to
polyethylene bags (with no air), where they remained during the following 48 h of fermentation.
To all samples, fermentation was then carried at 35 ºC, at atmospheric pressure (0.1 MPa), for 72
h. Fermentation samples were collected over time in duplicate and all the analyses were also
performed in duplicate.
In a subsequent study, fermentation was performed in polyethylene bags with air, under
two slightly different Vmedium:Vair ratios (1.0:1.8 and 1.0:2.2), to test the most suitable conditions
for fermentation under pressure, considering volume limitations of the high pressure vessel.
Fermentation was carried at 35 ºC, at atmospheric pressure (0.1 MPa), for 24 h. Fermentation
samples were collected over time in duplicate and all the analyses were also performed in
duplicate.
7.2.4. Fermentation experiments: Effect of high pressure

Fermentation was carried out in polyethylene bags, with controlled V medium:Vair ratio
(1.0:2.2), at 35 ºC under different HP conditions (10, 25, and 35 MPa), for 72 h. The experiments
were conducted in a Hydrostatic press (FPG7100, Stanstead Fluid Power, Stanstead, United
123
Kingdom), with a pressure vessel of 100 mm inner diameter and 250 mm height surrounded by
an external jacket to control the temperature, using a mixture of propylene glycol and water (40:60
v/v) as pressurizing fluid. In parallel, a control sample (at atmospheric pressure, 0.1 MPa) was
also performed, maintaining the exact same conditions of the HP-samples. Fermentation samples
were collected over time in duplicate and all the analyses were also performed in duplicate.
7.2.5. Analysis of biomass concentration

Waltham, Massachusetts, USA). Cell dry weight (CDW) was routinely determined using a
standard curve relating P. denitrificans optical density and cell dry weight (CDW).
7.2.6. Glycerol quantification

Glycerol measurement was performed in the samples supernatants using the Glycerol GK
Assay Kit (Megazyme, Ireland), accordingly to the manufacturer’s instructions for use in 96-well
microplates. The absorbance was measured with a Multiskan GO Microplate Spectrophotometer
(Thermo Fisher Scientific Inc., USA). The results were further confirmed by analysis with high
performance liquid chromatography (HPLC) coupled with refraction index detector (HPLC-RI),
by the method described in the following section.
7.2.7. Characterization of the extracellular medium

Culture samples were centrifuged at 10,000 rpm and 4 ºC for 10 min and the collected
using a HPLC Knauer system equipped with Knauer K-2301 RI detector and a Aminex HPX-87H
cation exchange column (300 x 7.8 mm) (Bio-Rad Laboratories Pty Ltd, Hercules, CA, USA).
The mobile phase was 13 mM H2SO4, delivered at a flow rate of 0.6 mL min-1 and the column
maintained at 65 °C. Peaks were identified by their retention times and quantified using
calibration curves prepared with the respective standards.

samples.
124
7.3.1. Effect of air availability on Paracoccus denitrificans growth and fermentation at

atmospheric pressure
In the first part of this work, P. denitrificans metabolism was evaluated under different
conditions of air availability and agitation, in order to understand how they affected growth and
fermentation. For that purpose, three different approaches were tested: (1) in the absence of air
(or, at least, minimizing the air availability) and without agitation, i.e. samples “without air”; (2):
in the presence of air (Vmedium:Vair ratio of 1:5) and with agitation, i.e. samples “with air”; and (3):
in a mixed process, which corresponded to fermentation with air in the first 24 h, and without air
in the remaining time, i.e. samples “24 h with air + 48 h without air” (without adding fresh culture
medium). We intended to assess if P. denitrificans was able to maintain growth and activity under
all these conditions, and to analyze potential metabolic differences between them. The results for
variation of biomass and glycerol concentrations are indicated in Figure 7.1.
Figure 7.1. Concentrations of biomass (a) and glycerol (b) throughout fermentation, for different air
availability conditions: without air during the entire process (●); with air during the first 24 h, and without
air during the remaining 48 h (▲); or with air during the entire process (■).
In samples “without air”, cell growth (Figure 7.1.a) was nearly inexistent and substrate
consumption was low, with glycerol concentrations (Figure 7.1.b) varying from 18.95 g L-1 in the
beginning of fermentation to 17.90 g L-1 after 72 h. This indicates that P. denitrificans was
inhibited by the absence of aeration/agitation and confirms that oxygen was required for cell
growth, at least when using this culture medium and conditions. In literature, P. denitrificans
strains are reported to grow both aerobically and anaerobically (Beijerinck and Minkman, 1910),
but with specific requirements of culture media composition and culture conditions at each one
of these environments (Hahnke et al., 2014; Nokhal and Schlegel, 1983). In the present work, the
selected culture medium and conditions did not seem suitable for growth under low oxygen
125
environments. In fact, higher air availability and agitation conditions were more suitable for P.
denitrificans growth, since the samples “with air” showed pronounced cell growth and substrate
consumption over time. In the “24 h with air + 48 h without air” samples, cell growth and glycerol
consumption were similar to samples “with air” during the first 24 h, as expected, since the
fermentation was performed at the same conditions during that period. Afterwards, when
fermentation was carried out without air, cell and glycerol concentrations remained stable over
time, possibly due to inhibition of metabolic activity.
The effects of air availability were also evaluated in terms of production of extracellular
compounds by P. denitrificans during fermentation. This is a rather unusual approach, since
typically only P. denitrificans intracellular products are analyzed, in attempt to find biopolymers
(Kalaiyezhini and Ramachandran, 2015; Mothes et al., 2007). However, since that was not the
purpose of the present work, only the extracellular products were analyzed, in particular alcohols
and organic acids. As a result, ethanol, acetate and succinate were identified for each fermentative
batch. The variation of these compounds over fermentation time is represented in Figure 7.2,
while the respective yields (Y, g g-1) are indicated in Table 7.1. The formation of the extracellular
products was profoundly affected by air availability. It is interesting to point out that the
conditions “without air”, which highly inhibited cell growth and glycerol uptake, were the ones
that promoted the formation of extracellular products.
Figure 7.2. Concentrations of ethanol (a), acetate (b) and succinate (c) throughout fermentation, for
different air availability conditions: without air during the entire process (●); with air during the first 24 h,
and without air during the remaining 48 h (▲); or with air during the entire process (■).
126
Table 7.1. Yields of biomass (YX/S), ethanol (YEtOH/S), acetate (YAcet/S), and succinate (YSucc/S) on glycerol,
for fermentation under different air availability conditions.
Samples YX/S (g g-1) YEtOH/S (g g-1) YAcet/S (g g-1) YSucc/S (g g-1)
Without air 0.061 ± 0.019 0.945 ± 0.005 0.140 ± 0.089 0.106 ± 0.018
24h with air + 48h

0.492 ± 0.005 0.095 ± 0.002 0.028 ± 0.001 n.d.
without air
With air 0.221 ± 0.008 n.d. n.d. n.d.
represent the mean ± SD of two independent biological replicates, analyzed in duplicated. N.d. indicates non-detected production of
the compound.
While all these compounds were formed in samples fermenting “without air”, none of them was
detected in samples with high air availability. In the mixed samples, ethanol and acetate (Figure
7.2.a and 7.2.b) were both produced, but only during the period of fermentation without air (24-
72 h). It would be expected that the higher cell density accumulated in mixed samples during the
first 24 h would result in increased production of extracellular compounds, compared to samples
“without air”. However, acetate concentration was lower in mixed samples, and succinate (Figure
7.2.c) was not produced, which suggests that both samples developed different mechanisms to
survive under low oxygen availability conditions. In addition, the absence of succinate may
indicate that this compound was produced by different metabolic pathways, which were
differently affected by air availability, relative to ethanol and acetate.
Since the extracellular products were only formed in samples “without air”, or during the
equivalent period in mixed samples, it is possible to conclude that these products are characteristic
of P. denitrificans metabolism under lower oxygen availability. A similar behavior was reported
for a recombinant E. coli strain (de Almeida et al., 2010). In that case, two different agitation
speeds were used to provide different levels of oxygen availability, and resulted in variations in
the pattern of product formation. In cultures grown with strong agitation, i.e. with higher oxygen
availability, there was low production of metabolic products (ethanol, and acetate, formate and
lactate) and formation of larger amounts of biomass. In contrast, the reduction in oxygen
availability caused a redirection of carbon flow towards the production of acids and ethanol. The
authors also observed that this enhancement effect was particularly noteworthy for ethanol
production, compared to the formation of organic acids. Similarly, in the present work, there was
high ethanol production during fermentation with low aeration (Figure 7.2 and Table 7.1), which
127
resulted in a concentration of 1.00 g L-1 and a yield of 0.95 g g-1, both considerably higher than
the values obtained for acetate (0.15 g L-1 and 0.14 g g-1) after the same time.
The high value obtained for ethanol yield on glycerol raised the possibility of production
of this compound (as well as other extracellular compounds) from carbon sources other than
glycerol. In fact, the experiments were performed using complex culture medium, since some
ingredients had unspecified chemical composition, e.g. yeast or meat extracts. Therefore, it would
be possible to have different carbon sources present in the media. In order to address this issue,
the presence of alternative carbon sources was evaluated, and glucose and maltose were detected
in the initial samples (0 h), as indicated in Table B.1 (Appendix B). Glucose concentration showed
low variation over time, from 1.07 g L-1 at 0 h, to 1.03 - 1.66 g L-1 after 72 h. In contrast, maltose
initially present in the medium (0.26 g L-1) was entirely consumed after 48 h of fermentation.
These results indicate that ethanol, acetate and succinate may also be produced from maltose, and
not exclusively from glycerol. To understand which substrate is used for the production of each
compound, and which are the metabolic pathways used for that purpose, the metabolic profile of
P. denitrificans should be studied in detail, using specific and suitable metabolomics tools.
In samples “without air”, P. denitrificans showed metabolic activity, with the production
of ethanol, acetate and succinate, but no cell growth over time. In contrast, samples “with air”
showed considerable cell growth, but no production of ethanol, acetate or succinate. Samples “24
h with air + 48 h without air” were able to accumulate biomass during the first 24 h, and to produce
ethanol and acetate during the period without air. However, it did not achieve the concentrations
produced without air, and succinate was not even detected. Overall, the results in this section
showed that P. denitrificans growth and metabolism were highly affected by air availability. It is
also important to consider that the low cell growth under low air availability can represent a
serious limitation to achieve reasonable concentrations of fermentation products, due to the
reduced number of cells. Therefore, the use of moderate air availability conditions would possibly
favor the process, balancing the formation of biomass and the fermentative activity. In addition,
these would be the most suitable conditions to perform fermentation under high pressure (HP),
due to volume limitations of the HP vessel used at the present work, which does not easily allow
high air volumes or agitation. Considering these constrains, two different Vmedium:Vair ratios
(1.0:1.8 and 1.0:2.2) were selected, and tested for P. denitrificans growth and fermentation during
24 h. As indicated in Table 7.2, specific growth rates (µ, h-1), final biomass concentrations and
glycerol consumption percentages were all slightly higher for the 1.0:2.2 ratio, correspondent to
higher air availability. It is certain that high air availability and high specific growth rates do not
only always translate in higher product formation (de Almeida et al., 2010; Kalaiyezhini and
Ramachandran, 2015). However, since the study under HP would be limited by other factors, such
128
as the lack of agitation, and the stress induced by the pressure itself, the highest ratio (1.0:2.2)
was selected to proceed the studies with P. denitrificans at HP conditions.
Table 7.2. Specific growth rates (µ), final biomass concentrations and percentages of glycerol consumed,
after 24 h of fermentation, for medium:air ratios of 1.0:1.8 and 1.0:2.2.
Final biomass
Samples µ (h-1) Glycerol consumed (%)
concentration (g L-1)
1.0:1.8 0.098 ± 0.011 2.65 ± 0.14 28.34 ± 3.19

0.1 MPa
1.0:2.2 0.104 ± 0.010 2.99 ± 0.16 33.54 ± 0.98
Values reported in the table represent the mean ± SD of two independent biological replicates, analyzed in duplicated.
7.3.2. Effect of high pressure on Paracoccus denitrificans growth and fermentation

The previously discussed experiments provided relevant information about the
fermentation process and how it depends on air availability. In the second stage of the work, HP
(at 10, 25 and 35 MPa) was applied during the 72 h at 35 ºC of P. denitrificans fermentation.
Fermentation was also tested at atmospheric pressure (0.1 MPa), to use as control. The pressure
effects on cell growth and glycerol consumption (Figure 7.3) showed a clear inhibitory effect,
which was more accentuated with the increase of pressure level.
Figure 7.3. Concentrations of biomass (a) and glycerol (b) throughout fermentation at different HP
conditions: 10 MPa (▲), 25 MPa (●), or 35 MPa (■). Control samples (0.1 MPa) are also represented (*).
Similar cell growth profiles (Figure 7.3.a) were observed at 10 and 25 MPa, with similar
biomass concentrations (p > 0.05) reached after 72 h of fermentation (2.52 and 2.42 g L -1,
respectively). In both cases, biomass concentration was significantly lower (p < 0.05) compared
to the obtained at 0.1 MPa (3.01 g L-1). At 35 MPa, inhibition of cell growth was even more
pronounced, showing only slight variation over time, which resulted in a final biomass
129
concentration of only 1.14 g L-1. In a study concerning microbial growth under hyperaccelerations
(in centrifuges), Deguchi et al. (2011) tested the effects of HP (being the pressure generated by
hyperacceleration) on P. denitrificans proliferation, and observed that the strain was able to grow
at 30 MPa, but was completely inhibited at 40 MPa (at 30 ºC, in LB agar). This negative effect of
HP on cell growth was previously reported for other microorganisms, such as Streptococcus
thermophilus, Lactobacillus bulgaricus, and Bifidobacterium lactis, at 5 and 100 MPa (Mota et
al., 2015); for Gluconacetobacter xylinus at 100 MPa (Kato et al., 2007); and for Clostridium
thermocellum at 7 and 17.3 MPa (Bothun et al., 2004).
The inhibitory effect of pressure on cell growth may result from a wide variety of
damaging effects. Generally, low pressure levels, such as the ones used in this work, may be
enough to impair several cellular processes, such as motility, cell division, nutrient uptake or
membrane protein function. A pressure of 50 MPa can inhibit protein synthesis and reduce the
number of ribosomes, while 100 MPa can induce partial protein denaturation (Abe, 2007; Huang
et al., 2014). However, it is important to take into account that pressure effects on microbial
growth are highly variable according to several factors, such as the organisms’ degree of
piezotolerance, the growth stage, the extent and duration of pressure treatment, as well as other
environmental parameters (Mota et al., 2013).
Glycerol consumption (Figure 7.3.b) seemed to be less affected by HP, at least at 10 and
25 MPa: in those cases, glycerol consumption after 48 h was slightly but significantly lower (p <
0.05) compared to 0.1 MPa, but reached similar concentrations (p > 0.05) after 72 h (in the range
of 9.04 and 9.26 g L-1). In contrast, biomass concentration at these same pressure conditions was
always lower (p < 0.05) compared to 0.1 MPa. This discrepancy between the pressure effects on
growth and substrate consumption may be related to the development of stress response
mechanisms to ensure cell survival. Under stress conditions, cell growth is usually disregarded,
in order to favor other processes more relevant to their survival, i.e. cell maintenance processes.
Cell maintenance refers to the fraction of substrate consumed to generate energy for functions
other than the production of new cell material (Pirt, 1965). These functions include energy costs
of osmoregulation, cell motility, turnover of macromolecular compounds, as well as defense
mechanisms (Van Bodegom, 2007). In short, when the energy is used for these maintenance
processes, bacterial growth is reduced, even if substrate consumption is maintained. Therefore, it
is expectable that under stressful conditions, such as HP, biomass production will be more
affected than substrate consumption. However, this effect was not observed at 35 MPa, which
showed low glycerol consumption during the entire process, possibly indicating metabolic
inhibition at this pressure.
As discussed in the previous section, glucose and maltose can both be found in the culture
medium, and its presence (and consumption) may have an impact on P. denitrificans metabolism.
130
Therefore, the concentrations of these sugars were analyzed throughout the fermentation time, at
the end of each fermentation condition. In all cases, maltose showed the same behavior, with the
0.31 g L-1 of maltose initially present in the medium being completely consumed, regardless of
the pressure applied (data not shown). On the other hand, glucose consumption was highly
affected by pressure, as indicated in Figure 7.4.
Figure 7.4. Glucose concentrations throughout fermentation at different HP conditions: 10 MPa (▲), 25
MPa (●), or 35 MPa (■). Control samples (0.1 MPa) are also represented (*).
After 48 h, glucose concentrations were similar (p > 0.05) at 0.1, 10 and 25 MPa (≈ 1.38
g L-1), but considerably changed at 72 h: i) at 0.1 MPa, the concentration remained as 1.38 g L -1;
ii) at 10 MPa, it decreased to 0.76 g L-1; iii) at 25 MPa, it decreased to 0.97 g L-1. This suggests
that HP stimulates glucose consumption during fermentation, possibly due to the higher need of
substrate and energy to ensure cell survival, with development of general and/or specific stress
responses. Interestingly, glucose concentrations at 35 MPa were significantly lower (p < 0.05)
that for all other conditions, after 48 and 72 h (0.83 and 0.20 g L -1, respectively), showing an
opposite behavior relative to glycerol consumption. Therefore, fermentation at 35 MPa seems to
stimulate glucose consumption, while the same effect was not observed for glycerol. These results
suggest specific metabolic changes at this pressure level, which affected differently P.
denitrificans growth and fermentation compared to the lower pressure levels.
Application of HP on P. denitrificans was also evaluated in terms of ethanol, acetate and
succinate production (Figure 7.5). Different pressure levels showed different effects on the
formation of these compounds. Ethanol production (Figure 7.5.a) was observed for all pressure
conditions, with a general increasing trend for the first 48 h of fermentation, and decreasing
thereafter.
131
The highest ethanol concentration was reached after 48 h at 0.1 MPa (1.92 g L-1), but this
compound was not detected after 72 h at this pressure. Similarly, ethanol concentrations of 1.34
and 1.50 g L-1 were observed after 48 h at 10 and 25 MPa, respectively, but these values decreased
to 0.43 and 0.40 g L-1 at 72 h of fermentation. These results suggest that ethanol formed during
the first hours of fermentation was then converted into other products. An option for the ethanol
degradation pathway is the oxidation into acetaldehyde, which can be followed by oxidation into
acetate.
Figure 7.5. Concentrations of ethanol (a), acetate (b) and succinate (c) throughout fermentation at different
HP conditions: 10 MPa (▲), 25 MPa (●), or 35 MPa (■). Control samples (0.1 MPa) are also represented
(*).
In fact, Felux et al. (2013) indicated that P. denitrificans Pd1222 (a derivative of DSM 413) has
the genetic machinery to perform these metabolic reactions: a gene that encodes an alcohol
dehydrogenase (locus tag Pden_2367) able to convert ethanol into acetaldehyde; and a gene that
encodes an NAD+-dependent aldehyde dehydrogenase (locus tag Pden_2366) that oxidizes
acetaldehyde to acetate. This is also supported by the production of acetate (Figure 7.5.b) during
the period between 48 and 72 h, with the highest concentration at 0.1 MPa (0.69 g L-1), followed
by significantly lower concentrations (p < 0.05) at 10 and 25 MPa (0.16 and 0.17 g L-1,
respectively). However, these acetate concentrations are too low relative to the concentration of
ethanol consumed. If ethanol was entirely converted into acetate, ≈ 2.5 g L-1 would be obtained
at 0.1 MPa, which is quite lower than the concentration actually detected (0.69 g L-1). There may
132
be two possible explanations for this discrepancy: i) acetaldehyde (obtained from ethanol) was
not entirely converted into acetate and accumulated in the cell, which is unlikely due to the high
toxicity of this compound; ii) acetate obtained from this pathway was further converted into
acetyl-CoA by an acetyl-CoA synthetase, possibly entering in a wide variety of metabolic
pathways, such as the tricarboxylic acid cycle (TCA cycle) or fatty acid synthesis.
Succinate production (Figure 7.5.c) was also detected during fermentation at 0.1, 10 and
25 MPa, but not at 35 MPa. After 72 h of fermentation, succinate concentrations were
significantly different (p < 0.05) for all pressures tested: the highest succinate concentration was
achieved for samples at 10 MPa (0.28 g L-1), followed by the ones at 0.1 MPa (0.21 g L-1) and,
finally, at 25 MPa (0.13 g L-1). Stimulation of succinate production at 10 MPa could be an
interesting outcome of P. denitrificans fermentation under HP, since this compound is widely
used as a precursor of many industrially important compounds in food, chemical, and
pharmaceutical industries (Jiang et al., 2017). However, the concentrations produced by P.
denitrificans are considerably low compared to other microorganisms typically used for that
purpose, such as Actinobacillus succinogenes, Mannheimia succiniciproducens, or
Anaerobiospirillum succiniciproducens. For those microbial strains, succinate concentrations in
the range of ≈ 10 – 83 g L-1 are usually reported in literature (Jiang et al., 2017). However, further
optimization is highly likely to be possible, and could increase the succinate production by P.
denitrificans, possibly resulting in titers and yields more similar to the ones reported for other
microorganisms.
As observed for other results in this section, fermentation at 35 MPa exhibited a different
metabolic profile compared to other pressure conditions, with low ethanol production (max. 0.17
g L-1), and no detected production of acetic and succinate. This suggests that HP is inhibiting the
formation of these compounds, an effect that was also observed at lower extent for fermentation
at 10 and 25 MPa. To clarify this inhibitory effect of pressure, the yields (Y, g g-1) of biomass,
ethanol, acetate, and succinate on glycerol were estimated at the end of fermentation (72 h), and
are indicated in Table 7.3. In the cases of biomass and acetate, the yields followed a decreasing
trend with the increase of pressure, suggesting a negative impact on these features. Ethanol yields
were only estimated for fermentation at 10 and 25 MPa, as this compound was not detected after
72 h at the other conditions. Due to the high variation of ethanol concentrations over time, the
yields at the end of fermentation did not allow a pertinent evaluation of the HP effects on ethanol
formation. In contrast, succinate yields reflected the behavior observed for concentrations over
time, with the yield at 10 MPa (0.29 g g-1) being slightly higher than at 0.1 MPa (0.22 g g-1).
133
Table 7.3. Yields of biomass (YX/S), ethanol (YEtOH/S), acetate (YAcet/S), and succinate (YSucc/Gly) on glycerol,
for fermentation under different pressure conditions.
Pressure
YX/S (g g-1) YEtOH/S (g g-1) YAcet/S (g g-1) YSucc/S (g g-1)
(MPa)
0.1 0.221 ± 0.008 n.d. 0.070 ± 0.006 0.022 ± 0.003
10 0.170 ± 0.012 0.043 ± 0.007 0.016 ± 0.003 0.029 ± 0.005
25 0.163 ± 0.010 0.041 ± 0.011 0.017 ± 0.001 0.014 ± 0.005
35 0.121 ± 0.056 n.d. n.d. n.d.
represent the mean ± SD of two independent biological replicates, and analyzed in duplicated. N.d. indicates non-detected production
of the compound.
Overall, HP was found to affect P. denitrificans cell growth and metabolism, with
different effects on substrate consumption, as well as on production of ethanol, acetic and
succinate. These effects varied according to the pressure level, with the lower pressures (10 and
25 MPa) showing a behavior approximate to 0.1 MPa, while the highest pressure (35 MPa)
presented a more extensive impact on P. denitrificans metabolism. Such as previously reported
for other microorganisms (Mota et al., 2018), the application of pressure stresses resulted in
particular and interesting effects on P. denitrificans growth and metabolism. In this preliminary
study, the experiments were performed in a lab-scale HP equipment, designed for pasteurization
and food technology purposes, that can also be used for a broader range of applications, including
extraction, hyperbaric storage, or microbial growth, but with inherent limitations. In the case of
microbial growth and fermentation processes under pressure, the main constrains are related to
volume limitations, absence of agitation mechanisms, as well as unpractical oxygen supply.
Therefore, it may be worth to perform equipment and process optimization, in order to perform
further studies in more suitable and tailor-made systems, able to meet the specifications of these
microbial processes. In fact, such type of pressure equipment is now becoming more widely
available, making it possible to evaluate the full potential of fermentation under pressure. It should
be highlighted that, in this context, this technology is highly versatile, since it can be applied
intermittently, as pressure stresses, but can also be maintained during the whole fermentation
time, without serious cell loss and no heating effect. Since there is no refrigeration requirement
(because the continuous application of pressure does not generate heat), the energetic costs of the
fermentation process are lower, and the application of HP to these processes is simpler.
134
Additionally, it is only necessary to provide energy to generate the pressure (and not to maintain
it), and so application of HP stress during the whole fermentation process has minimal energetic
costs, which would have a small impact on the integration of high pressure on industrial
fermentative processes. Also, as the pressure levels used in these processes are quite lower than
those used for food processing, the required equipment could be designed to withstand lower
pressures, thus being cheaper than the commercial alternatives currently on the market. Therefore,
HP technology can offer a high variety of process possibilities to perform microbial growth and
fermentation under pressure.
7.4. Conclusions
The present work intended to study the possibility of applying HP to P. denitrificans
glycerol fermentation, to stimulate cell growth and/or improve fermentation. However, it was
necessary to consider that some of HP systems that may be used for these purposes currently
comprise some limitations to aerobic processes, such as the absence of continuous air supply or
agitation. To understand if it was possible to perform P. denitrificans growth and fermentation
under limited-air conditions, the effects of air availability on this process were evaluated. The
results showed that growth and metabolism were both highly affected by air availability. With
higher air availability, considerable cell growth was observed over time, but no production of
ethanol, acetate or succinate. In contrast, without air availability, P. denitrificans showed active
metabolic activity (with the production of ethanol, acetate and succinate), but no cell growth over
time. Therefore, these products seem to be characteristic of P. denitrificans metabolism under
lower oxygen availability.
To avoid inhibition of both cell growth and formation of extracellular products,
fermentation at HP conditions was tested under moderate air availability conditions (Vmedium:Vair
ratio of 1.0:2.2). Paracoccus denitrificans cells were able to grow under HP, even if at lower
extent compared to atmospheric pressure. At 10 and 25 MPa, biomass concentrations were still
similar to 0.1 MPa, while a more extensive inhibitory effect was observed at 35 MPa. In fact, this
pressure may be enough to impair several cellular processes, resulting in decreased cell growth
under these conditions. Application of HP was also found to promote modifications in terms of
substrate consumption, and formation of ethanol, acetate and succinate, with the fermentative
profile varying according to the pressure level. Generally, it was similar at 10 and 25 MPa, but
considerably different at 35 MPa, possibly as a result of metabolic shifts, or even inhibition. The
formation of these compounds under HP showed interesting patterns, and confirm that HP has
interesting effects on living systems, offering great biotechnological potential (such an example
are microorganisms thriving in deep-sea). Therefore, it would be interesting to proceed the studies
135
on P. denitrificans under HP, to further optimize the fermentation process at these conditions, and
to improve the titers and yields. As future work, it would also be relevant to study the pressure
effects on the production of different compounds, such as the intracellular biopolymers
polyhydroxyalkanoates. It would also be important to evaluate in more detail the effects of HP on
cell growth and viability, in order to estimate relevant kinetic parameters (that can only be
calculated by collecting more samples over time, and thus adding more data points) and modelling
these fermentation processes.
Overall, the implications of the pressure-promoted changes in P. denitrificans growth and
fermentation process are still not completely understood, but the results obtained in this work
unveil new metabolic features of this bacterial strain, and provide useful information for further
studies regarding P. denitrificans under pressure. From a more comprehensive view, this study
also opens the way for application of this technology to other glycerol fermentation processes, in
particular to the ones with high requirements of air availability. It is certain that, in these particular
cases, there are more limitations of the process (mostly due to volume constrains) that demand a
more complex optimization, but the results obtained so far for P. denitrificans confirm that is
possible to perform aerobic growth under these lower air availability conditions.
136
7.5. References
Abe, F., 2007. Exploration of the effects of high hydrostatic pressure on microbial growth,
physiology and survival: Perspectives from piezophysiology. Biosci. Biotechnol. Biochem.
71, 2347–2357.
Beijerinck, M.W., Minkman, D.C.J., 1910. Bildung und verbrauch von stickoxydul durch
bakterien. Zentbl. Bakteriol. Parasitenkd. Infekt. Hyg. Abt. II 25, 30–63.
de Almeida, A., Giordano, A.M., Nikel, P.I., Pettinari, M.J., 2010. Effects of aeration on the
synthesis of poly (3-hydroxybutyrate) from glycerol and glucose in recombinant
Escherichia coli. Appl. Environ. Microbiol. 76, 2036–2040.
7997–8002.
Felux, A.-K., Denger, K., Weiss, M., Cook, A.M., Schleheck, D., 2013. Paracoccus denitrificans
PD1222 utilizes hypotaurine via transamination followed by spontaneous desulfination to
yield acetaldehyde and, finally, acetate for growth. J. Bacteriol. 195, 2921–2930.
Follonier, S., Henes, B., Panke, S., Zinn, M., 2012. Putting cells under pressure: A simple and
efficient way to enhance the productivity of medium-chain-length polyhydroxyalkanoate in
processes with Pseudomonas putida KT2440. Biotechnol. Bioeng. 109, 451–461.
Garcia-Ochoa, F., Gomez, E., 2009. Bioreactor scale-up and oxygen transfer rate in microbial
processes: An overview. Biotechnol. Adv. 27, 153–176.
Hahnke, S.M., Moosmann, P., Erb, T.J., Strous, M., 2014. An improved medium for the anaerobic
growth of Paracoccus denitrificans Pd1222. Front. Microbiol. 5, 18.
Hori, K., Soga, K., Doi, Y., 1994. Effects of culture conditions on molecular weights of poly (3-
hydroxyalkanoates) produced by Pseudomonas putida from octanoate. Biotechnol. Lett. 16,
709–714.
Jiang, M., Ma, J., Wu, M., Liu, R., Liang, L., Xin, F., Zhang, W., Jia, H., Dong, W., 2017. Progress
of succinic acid production from renewable resources: Metabolic and fermentative
strategies. Bioresour. Technol. 245, 1710–1717.
137

698.
Eng. Life Sci. 7, 475–479.
426–433.
Nokhal, T.-H., Schlegel, H.G., 1983. Taxonomic study of Paracoccus denitrificans. Int. J. Syst.
Evol. Microbiol. 33, 26–37.
Pirt, S.J., 1965. The maintenance energy of bacteria in growing cultures, in: Proceedings of the
Royal Society B. pp. 224–231.
3574–3579.
Van Bodegom, P., 2007. Microbial maintenance: A critical review on its quantification. Microb.
Ecol. 53, 513–523.
138
139
140
CHAPTER VIII
Effect of high pressure on

Paracoccus denitrificans growth and
polyhydroxyalkanoates production
Adapted from:
Mota, M.J., Lopes, R.P., Simões, M.M.Q., Delgadillo, I., Saraiva, J.A., Effect of high pressure on
Paracoccus denitrificans growth and polyhydroxyalkanoates production from glycerol.
Submitted in Journal of Industrial Microbiology & Biotechnology.
CHAPTER VIII. Effect of high pressure on P. denitrificans growth and polyhydroxyalkanoates production
8.1. Introduction
Biodiesel corresponds to a fuel originated from biomass, and represents one promising
alternative to fossil fuels (da Silva et al., 2009). One of the major problems of biodiesel production
is the formation of glycerol as a reaction by-product (da Silva et al., 2009; Kolesárová et al.,
2011). In consequence, for the effective utilization of the excess glycerol, it became necessary to
develop new and sustainable applications, such as the use as substrate for conversion into high-
value bioproducts. In the last decades, significant research efforts have been focusing on this
subject, aiming the production of several valuable products, including alcohols, organic acids (e.g.
citrate, succinate), polymers, among others (Mattam et al., 2013).
Paracoccus denitrificans is a Gram-negative facultative methylotrophic bacterium able
to grow in glycerol, as well as in many other carbon sources, including methanol, ethanol, butan-
1-ol, and pentan-1-ol (Ueda et al., 1992; Yamane et al., 1996a, 1996b). It exhibits metabolic
versatility, and it was shown to grow aerobically and anaerobically. Paracoccus denitrificans is
reported in literature as a producer of polyhydroxyalkanoates (PHA), a complex class of naturally
occurring bacterial polyesters (Ashby et al., 2004). Paracoccus denitrificans was found to
accumulate poly(3-hydroxybutyrate-co-3-hydroxyvalerate) during growth on pentan-1-ol, with
the polymer compositions varying during the fermentation time (Yamane et al., 1996b).
Paracoccus denitrificans also synthesized co-polymer of poly(3-hydroxybutyrate-co-3-
hydroxyvalerate) when methanol and an amyl alcohol were added together to a nitrogen-limited
medium (Ueda et al., 1992). Mothes et al. (2007) firstly reported the production of poly(3-
hydroxybutyrate) by P. denitrificans using crude glycerol as carbon source. Kalaiyezhini and
Ramachandran (2015) have also studied poly(3-hydroxybutyrate) production from glycerol by P.
denitrificans. In that case, the kinetics of poly(3-hydroxybutyrate) biosynthesis was evaluated in
a batch bioreactor, testing different operational parameters, such as nitrogen source, carbon to
nitrogen ratio, pH, aeration, and initial glycerol concentration, and the authors observed that the
most suitable conditions for bacterial growth were not the same as for PHA production and
accumulation.
The performance of fermentation under non-conventional conditions (such as high
pressure, electric fields or ultrasounds) is a strategy that is being currently tested for different
fermentation processes (Mota et al., 2018). When these stresses are applied at sub-lethal levels,
the microbial strains may develop specific genetic, physiologic and metabolic responses,
promoting modification of fermentation products and processes. In some cases, these
modifications can represent considerable improvements, such as increased yields, productivities
and fermentation rates, lower accumulation of by-products and/or production of different
compounds. Therefore, the application of these non-conventional conditions to glycerol-based
fermentations could introduce significant improvements in these processes, making them more
143
attractive for industrial production. High pressure (HP) is one of the technologies used for this
purpose, with interesting results obtained so far, such as reviewed by Mota et al. (2013). The
application of HP allows more possibilities than other technologies (electric fields, ultrasounds)
since it can be applied as intermittent pressure stress, or can also be maintained during the whole
fermentation time, without serious cell loss and no heating effect. Since there is no refrigeration
requirement, the energetic costs of the fermentation process are lower, and the application of HP
to these processes is simpler. Additionally, it is only necessary to provide energy to generate the
pressure (and not to maintain it), and so application of HP stress during the whole fermentation
process has minimal energetic costs. Therefore, the whole fermentative processes can be easily
performed under HP stress conditions, differently from other technologies. This new feature
might be an advantage for using HP as a non-conventional technology for fermentative processes,
while the main drawback is still the high equipment costs.
One of the studies evaluating HP to fermentation used the yeast Saccharomyces
cerevisiae as a case-study of alcoholic fermentation under HP (5-100 MPa) and observed that
fermentation proceeded faster at pressures up to 10 MPa, compared to atmospheric pressure
(Picard et al., 2007). Application of HP throughout fermentation can also be used to change
product selectivity during the process. For instance, Bothun et al. (2004) applied pressures of 7
and 17.3 MPa on Clostridium thermocellum, and observed a shift in product selectivity from
acetate to ethanol. At the applied pressures, ethanol:acetate ratio had a 60-fold increase relatively
to atmospheric pressure. Another possibility is the application of HP to modify the properties of
biopolymers produced during fermentation. For instance, cellulose produced by
Gluconacetobacter xylinus under HP conditions (30, 60 and 100 MPa) showed several
morphological changes (including higher density), compared to the polymer produced at
atmospheric pressure. This modification of cellulose morphological properties may promote the
acquisition of different functional properties (Kato et al., 2007). Regarding polymer production,
Follonier et al. (2012) applied a low pressure level (0.7 MPa) to Pseudomonas putida KT2440
and were able to enhance productivity of medium-chain-length polyhydroxyalkanoate production,
even with a significant decrease in specific growth rates. The effects of HP have also been
evaluated in the context of food fermentation: on lactic acid fermentation, for production of
probiotic yogurt (Mota et al., 2015); and in the beginning of malolactic fermentation by
Oenococcus oeni (Neto et al., 2016).
Currently, there is only one study on literature approaching P. denitrificans growth under
HP (Deguchi et al., 2011). In this work, a variety of microorganisms were cultured in nutrient
media under hyperaccelerations (in centrifuges), to evaluate its ability to grow under these
conditions. For instance, in the case of P. denitrificans, the cells were able to proliferate even at
403,627 × g. In order to study the role of HP on P. denitrificans, the authors evaluated the growth
144
of this microbial strain at 0.1, 30 and 40 MPa, at 30 ºC, being the pressure generated by
hyperacceleration. As a result, they observed that P. denitrificans retained the ability to grow at
30 MPa, while it was completely inhibited at 40 MPa. These are interesting but preliminary results
regarding the behavior of P. denitrificans in response to HP stress. Since this microbial strain has
high biotechnological potential and applicability, the present work intended to evaluate the effects
of different pressure levels (0.1 - 50 MPa) on P. denitrificans growth and fermentation, with
special focus on the formation of PHA.

A lyophilized culture of Paracoccus denitrificans DSMZ 413, obtained from DSMZ,
Germany, was used in this study. The strain was reconstituted on nutrient broth according to the
manufacturer’s instructions. The strain was sub-cultured on nutrient agar plates and incubated at
30 ºC for 24 h and then preserved at 4 ºC for a maximum period of 1 month.
Rich Medium reported by Hori et al. (1994) was used for inoculum preparation. It
included polypeptone (10 g L-1), yeast extract (10 g L-1), meat extract (5 g L-1) and (NH4)2SO4 (5
g L-1). Mineral medium reported by Mothes et al. (2007) was used for the fermentation
experiments. It contained glycerol (20 g L-1), yeast extract (4.5 g L-1), K2HPO4 (5 g L-1), KH2PO4
(0.5 g L-1), CaCl2·2H2O (20 mg L-1), MgSO4·7H2O (1 g L-1), and trace elements solution (2 mL
L-1) for PHB production. The composition of trace elements solution is as follows: FeSO 4·7H2O
(4.98 g L-1), ZnCl2 (0.44 g L-1), CuSO4·5H2O (0.78 g L-1), Na2MoO4·2H2O (0.24 g L-1),
MnSO4·4H2O (0.81 g L-1), dissolved in 1 N HCl solution.

A single cell colony was seeded into 100 mL of rich medium and incubated overnight at
35 ºC for 16 – 20 h, in a shaker (160 rpm).

Mineral medium was inoculated with 5 % (v/v) of standard inoculum. The mixture was
homogenized and then transferred to polyethylene bags, with defined air volume: in most cases,
with Vmedium:Vair ratio of 1.0:2.2, and in others with Vmedium:Vair ratio of 1:5, as detailed below for
each set of samples. All the preparation steps were performed in an aseptic environment, within
a laminar flow cabinet, to avoid sample contamination. Fermentation was carried at 35 ºC under
different HP conditions (10, 25, 35 and 50 MPa) for 72 h. The experiments were conducted in a
145
Hydrostatic press (FPG7100, Stanstead Fluid Power, Stanstead, United Kingdom), with a
pressure vessel of 100 mm inner diameter and 250 mm height surrounded by an external jacket
to control the temperature, using a mixture of propylene glycol and water as pressurizing fluid. In
parallel, two different control samples were used in this work: one of them at shake-flasks, with
Vmedium:Vair ratio of 1:5, and agitation speed of 135 rpm – C1; the other packed at conditions
similar to the HP-samples (i.e. in polyethylene bags, with Vmedium:Vair ratio of 1.0:2.2) – C2. The
former intended to be a control with high air availability and agitation, while the latter intended
to be the control for the experiments under pressure. Both control samples were incubated at
atmospheric pressure (0.1 MPa) and 35 ºC. To all pressure conditions (including atmospheric
pressure), fermentation experiments were performed in duplicate. Samples were collected over
time, and the analyses were also performed in duplicate.

USA). Cell dry weight (CDW) was routinely determined using a standard curve relating P.
denitrificans optical density and cell dry weight (CDW).

samples were prepared and aliquots of 1.0 mL of proper dilutions were plated in nutrient agar
plates, incubated at 30 °C for 24 h.
8.2.6. Glycerol quantification

Glycerol quantification was performed in the samples supernatants using the Glycerol
GK Assay Kit (Megazyme, Ireland), accordingly to the manufacturer’s instructions for use in 96-
well microplates. The absorbance was measured with a Multiskan GO Microplate
Spectrophotometer (Thermo Fisher Scientific Inc., USA).
8.2.7. Analysis of polyhydroxyalkanoate formation

After centrifugation, cell pellets were re-suspended in saline solution (0.9 % NaCl),
frozen and lyophilized under vacuum for ≈ 24 h. Gas chromatography-based analytical methods
require the PHA to be depolymerized and chemically converted into methyl ester derivatives (by
methanolysis) prior to analysis. Therefore, a weighed amount (20 mg) of dry cells was combined
with 2 mL of acidified methanol (20 % H2SO4), and 2 ml of chloroform. Benzoic acid was used
as internal standard. For methanolysis, samples and standards were heated at 100 ºC for 3 h Pyrex
146
test tubes with Teflon-lined caps, to convert the constituents to their methyl esters (Braunegg et
al., 1978). After that time, 1 ml of water was added to the reaction mixture, followed by vigorous
shaking during 2 min, to induce phase separation. The chloroform layer containing the PHA
methyl esters was separated and analyzed by GC-MS (gas chromatography coupled with mass
spectrometry) using a Gas Chromatograph Mass Spectrometer (GC-MS Shimadzu QP2010 Ultra)
coupled to an AOC 20i autosampler (Shimadzu, Japan) and with the electron impact ionization
(EI) at 70 eV. 1 µL of sample was injected automatically with a 10 µL glass syringe into a VF-
5ms column (30 m x 0.25 mm with 0.25 µm film thickness; Varian, Inc., USA). Helium was used
as the carrier gas and the linear velocity was set at 40 cm s-1. The oven temperature conditions
were set at 60 ºC for 2 min, followed firstly by a rise to 150 ºC at a 5 ºC per minute rate, then at
25 ºC per minute until 200 ºC where it stabilized for 3 min. Data were evaluated using the NIST
14 Mass Spectral database, and identifications were validated after injection of standards.

The results obtained were tested at a 0.05 level of probability, and the effect of pressure
was tested with a one-way analysis of variance (ANOVA), followed by a multiple comparisons
test (Tukey HSD) to identify the differences between samples.

Different levels of pressure (0.1 – 50 MPa) were applied to P. denitrificans fermentation,
and were found to affect different aspects of this process, such as cell growth, substrate
consumption and polymer formation. In the first stage of the work, the pressure effects were
evaluated in terms of cell growth, through the analysis of biomass concentrations and viable cell
counts, as indicated in Figure 8.1. Biomass concentration was evaluated over time, at 10, 25, 35
and 50 MPa, and also at 0.1 MPa (C1 and C2). At 50 MPa, fermentation was interrupted after 48
h of fermentation, since biomass concentration remained stable at this pressure, indicating no
occurrence of cell growth. Deguchi et al. (2011) have also observed that P. denitrificans growth
was completely inhibited at 40 MPa, 30 ºC. At lower pressure conditions (in the range of 0.1 - 35
MPa), P. denitrificans was able to grow, at more or less extent. The highest biomass
concentrations were achieved at 0.1 MPa, in particular for the C1 control (the one with agitation
and higher air availability). However, both atmospheric pressure samples showed similar biomass
concentrations after 72 h (p > 0.05): 3.09 and 3.01 g L-1, for C1 and C2, respectively.
147
Figure 8.1. Biomass concentration over time and viable cell counts after 72 h, for fermentation at different
pressure conditions: 10, 25, 35 and 50 MPa. Control samples, corresponding to fermentation at 0.1 MPa,
are also indicated as C1 (higher air availability and agitation) and C2 (lower air availability and no
agitation).
Biomass growth was significantly lower (p < 0.05) under HP conditions, suggesting an inhibitory
effect of pressure on P. denitrificans, similarly to the previously observed for this microorganism
(Deguchi et al., 2011), and for other microbial strains (Bothun et al., 2004; Kato et al., 2007; Mota
et al., 2015).Biomass growth was significantly lower (p < 0.05) under HP conditions, suggesting
an inhibitory effect of pressure on P. denitrificans, similarly to the previously observed for this
microorganism (Deguchi et al., 2011), and for other microbial strains (Bothun et al., 2004; Kato
et al., 2007; Mota et al., 2015). At 10 and 25 MPa, similar cell growth profile was observed over
time, with biomass concentrations of 2.52 and 2.42 g L-1, respectively, after 72 h. By increasing
the pressure to 35 MPa, the inhibitory effect became more pronounced, with low biomass
concentrations during the entire fermentation time: after 72 h, a biomass concentration of only
1.14 g L-1 was reached. The specific growth rates (µ, h-1), in Table 8.1, showed a similar trend,
considerably decreasing with the increasing pressure, from 0.085 h-1 at 0.1 MPa/C2, to 0.040 h-1
at 10 and 25 MPa, and to 0.020 h-1 at 35 MPa.
Viable cell counts after 72 h showed a slightly different behavior relatively to biomass
concentration (Figure 8.1.b). For instance, cell counts at 10 MPa were closer to 0.1 MPa (both C1
and C2). At 10 and 25 MPa, P. denitrificans cells retained their ability to grow, from 7.46
log10(CFU mL-1) at the beginning of fermentation, to 9.07 and 7.98 log10(CFU mL-1), respectively.
In contrast, at 35 MPa, viable cell counts after 72 h were similar to those at the beginning of the
process (0 h), suggesting inhibition of cell growth by this pressure level, but not cell inactivation
(since there was no loss of viability).
148
Table 8.1. Specific growth rates (µ), percentages of glycerol consumed, and yields of biomass on glycerol,
after 72 h of fermentation, at different pressure conditions (0.1 – 35 MPa).
Samples µ (h-1) Glycerol consumed (%) YX/S
0.1 MPa (C1) 0.135 ± 0.002 58.3 ± 1.3 0.207 ± 0.001
0.1 MPa (C2) 0.085 ± 0.013 52.7 ± 2.0 0.221 ± 0.008
10 MPa 0.040 ± 0.005 51.7 ± 3.5 0.170 ± 0.012
25 MPa 0.040 ± 0.002 50.9 ± 2.8 0.163 ± 0.010
35 MPa 0.020 ± 0.002 12.1 ± 3.3 0.115 ± 0.002
represent the mean ± SD of two independent biological replicates, analyzed in duplicate.
The effects of HP on glycerol concentration over time are represented in Figure 8.2, while
the percentages of glycerol consumption are indicated in Table 8.1. As expected, substrate
consumption by P. denitrificans was affected by application of HP during fermentation, but at
lower extent compared to cell growth.
Figure 8.2. Glycerol concentrations over time, for fermentation at different pressure conditions: 10, 25, 35
and 50 MPa. Control samples, corresponding to fermentation at 0.1 MPa, are also indicated as C1 (higher
air availability and agitation) and C2 (lower air availability and no agitation).
149
During the first 48 h at 0.1 MPa, glycerol consumption was faster compared to other
pressure conditions, but the consumption decelerated (or even stopped) thereafter. In
consequence, at 72 h, glycerol consumption was similar (p > 0.05) between fermentations at 0.1
MPa/C2 (52.7 %), 10 MPa (51.7 %) and 25 MPa (50.9 %). Such as observed for cell growth,
substrate consumption at 35 MPa was considerably affected (p < 0.05), corresponding to only ≈
12 % after 72 h of fermentation. At 50 MPa, glycerol concentration showed almost no variation
over time (from 18.95 g L-1 at 0 h, to 17.99 g L-1 at 48 h), suggesting inhibition of P. denitrificans
metabolism at this pressure, such as observed for cell growth.
Yields of biomass on glycerol (YX/S, g g-1), indicated in Table 8.1, provide quantitative
information about the utilization of glycerol for cell growth. At atmospheric pressure, the yield at
0.1 MPa/C2 (0.221 g g-1) was slightly higher than at 0.1 MPa/C1 (0.207 g g-1), meaning that a
higher proportion of glycerol was being converted into biomass when fermentation was
performed under more limiting-oxygen conditions. For fermentation under HP, biomass yields
were always lower than at atmospheric pressure, and with a gradual decrease with the increase of
pressure: 0.170, 0.163 and 0.115 g g-1, for 10, 25 and 35 MPa, respectively. This indicates that,
under HP conditions, P. denitrificans tends to use a lower proportion of glycerol for biomass
production. A possible explanation for this effect relies on the utilization of glycerol for other
purposes different than cell growth, such as stress response mechanisms to ensure cell survival,
or other stress maintenance processes, which are certainly activated under HP conditions. This
can represent a positive feature for fermentation under pressure, since it may be related to the
increased production of different valuable metabolites. One class of these P. denitrificans
metabolites are polyhydroxyalkanoates (PHA), which play a pivotal role in priming
microorganisms for stress survival. It promotes the long-term survival of bacteria under nutrients-
scarce conditions by acting as carbon and energy reserves. In addition, bacteria that harbor PHA
showed enhanced stress tolerance against transient environmental assaults, such as ultraviolet
(UV) irradiation, heat and osmotic shock (Tan et al., 2014). Therefore, the effects of fermentation
under pressure on PHA production by P. denitrificans were evaluated, and the PHA contents,
yields (YPHA/S, g g-1) and productivities (QPHA, mg L-1 h-1), after 72 h of fermentation are indicated
in Table 8.2. The results for fermentation at 50 MPa were not included, since no PHA production
was detected after 48 h.
150
Table 8.2. Polyhydroxyalkanoate contents, yields and productivities, after 72 h of fermentation, at different
pressure conditions (0.1 – 35 MPa).
Samples PHA (g L-1) PHA (wt%) YPHA/S (g g-1) QPHA (mg L-1 h-1)
0.1 MPa (C1) 0.171 ± 0.011 5.5 0.016 ± 0.001 2.37 ± 0.15
0.1 MPa (C2) 0.120 ± 0.006 4.0 0.012 ± 0.000 1.65 ± 0.07
10 MPa 0.104 ± 0.018 4.1 0.011 ± 0.003 1.45 ± 0.25
25 MPa 0.105 ± 0.009 4.3 0.011 ± 0.000 1.45 ± 0.12
35 MPa 0.066 ± 0.009 5.8 0.026 ± 0.006 0.91 ± 0.01
represent the mean ± SD of two independent biological replicates, analyzed in duplicate.
The polymer concentrations achieved in this work (0.066 – 0.171 g L-1) are low, compared
to some other studies on PHA production by Paracoccus sp. strains, where optimized conditions
were used. For instance, Kalaiyezhini and Ramachandran (2015) and Kumar et al. (2018)
achieved PHA titers of 10.7 and 9.5 g L-1, respectively, in batch bioreactor experiments. However,
there are other studies about PHA production that report lower titers, in the range of those
observed in the present work (< 0.5 g L-1 (Davis et al., 2013; Kenny et al., 2008; Ueda et al.,
1992). Nevertheless, it is important to highlight that this work intended to study the effects of
pressure, and fermentation was carried out with a different and simpler experimental set-up, with
lower volumes more suitable for the HP experiments, due to the limited volume of the pressure
vessel and not under optimized conditions.
In the case of 0.1 MPa/C1, a PHA production of 0.171 g L-1 was detected, corresponding
to 5.5 % of cell dry mass. On the other hand, at 0.1 MPa/C2, PHA production decreased almost
30 % relatively to 0.1 MPa/C1, showing that lower oxygen availability conditions reduced PHA
production. This also reflects on PHA yields and productivities, both lower at 0.1 MPa/C2. These
results contrast with some of the studies in literature, which state that lower oxygen availability
stimulates polymer production and decreases cell growth. Kalaiyezhini and Ramachandran (2015)
observed that higher oxygen availability increased specific growth rates, while moderate oxygen
availability promoted PHA production. De Almeida et al. (2010) reported a similar behavior for
a recombinant Escherichia coli strain: higher oxygen availability resulted in lower production of
metabolic products (acids, ethanol and PHA) and formation of larger amounts of biomass;
151
reduction of oxygen availability caused a redirection of carbon flow towards the production of
those metabolic products. However, this effect might depend on several aspects, such as the extent
of the oxygen-limiting conditions.
Regarding the HP effects, there was a general trend to decrease polymer production with
the increasing pressure. Fermentation at 10 and 25 MPa (p > 0.05) showed similar PHA titers,
both of them ≈ 13 % lower than at 0.1 MPa/C2. Polyhydroxyalkanoate production was highly
affected at 35 MPa, resulting in a titer 45 % lower relatively to 0.1 MPa/C2. Interestingly, the
PHA content in cell dry mass (%) showed an opposite increasing trend with the increase of
pressure, indicating that, at HP conditions, P. denitrificans favored the polymer production
instead of biomass growth. The PHA content in the cells was particularly high for fermentation
at 35 MPa, which showed low biomass accumulation during the 72 h of fermentation. Similarly,
the PHA yield at 35 MPa was also high, and even considerably higher than the obtained for all
other fermentation conditions. At this pressure, glycerol consumption throughout the process was
rather low, but it seems that a high proportion of this glycerol was being directed to PHA
formation.
In sum, HP was found to decrease PHA production, resulting in lower titers and
productivities; however, the PHA content in cell dry mass tended to increase under pressure,
indicating that polymer production was being favored over biomass formation. Follonier et al.
(2012) reported a different behavior for Pseudomonas putida KT2440 at ≈ 0.7 MPa, with an
increase of PHA volumetric productivity under these pressure conditions. In that case, cell growth
was not even inhibited, under specific values of dissolved oxygen tension and dissolved carbon
dioxide tension. However, that pressure is considerably lower than the pressure levels used in the
present work, and, as a result, the extent of stress and damage inflicted to the cells is not
comparable.
Another possible effect of HP on PHA production may be related to modification of
polymer composition. Polyhydroxyalkanoates are diverse in their chemical composition and
material properties, due to the myriad of PHA monomeric units available, as well as the
incorporation of these monomers at varying amounts (Tan et al., 2014). Therefore, the monomeric
composition (mol%) of PHA produced under different pressure conditions was evaluated, and is
represented in Table 8.3. After 72 h at 0.1 MPa/C1, P. denitrificans accumulated a homopolymer
of 3-hydroxybutyrate, i.e. poly(3-hydroxybutyrate), which is in accordance with the results of
Kalaiyezhini and Ramachandran (2015) and Mothes et al. (2007) for PHA production by P.
denitrificans. Recently, Kumar et al. (2018) reported the production of PHA from glycerol by a
culture of Paracoccus sp. LL1, isolated from Lonar lake, India, and, in that case, production of a
poly(3-hydroxybutyrate-co-3-hydroxyvalerate) copolymer was observed. This was already
reported for other Paracoccus strains, using glucose, methanol, or pentan-1-ol as substrate
152
(Üçisik-Akkaya et al., 2009; Ueda et al., 1992; Yamane et al., 1996b). In the present work, the
production of the 3-hydroxyvalerate monomer was assessed with an appropriate standard, but it
was not detected in any of the samples.
In the case of fermentation with lower oxygen availability (0.1 MPa/C2, and HP
conditions), PHA composition was found to be rather different: the 3-hydroxybutyrate monomer
was not produced in these samples, and neither 3-hydroxyvalerate. Instead, medium-chain length
PHA (mcl-PHA; 6 to 14 carbon atoms) were detected in all cases. Some of these monomers were
identified, by comparing the mass spectra with NIST 14 MS database, together with the use of
appropriate standards. However, some monomers were not surely identified, and are thus
indicated as “unidentified” in Table 8.3.
Table 8.3. Polyhydroxyalkanoate monomeric composition (mol%), after 72 h of fermentation, at different

pressure conditions (0.1 – 35 MPa).
Samples 3-OH-C4 3-OH-C12 12-OH-C13 3-OH-C14 Unidentified
0.1 MPa (C1) 100 ± 0.0 n.d. n.d. n.d. n.d.
0.1 MPa (C2) n.d. 5.0 ± 1.0 50.2 ± 2.5 5.3 ± 2.0 39.5 ± 1.5
10 MPa n.d. 9.9 ± 0.7 5.7 ± 0.2 n.d. 84.4 ± 1.0
25 MPa n.d. 0.2 ± 0.0 50.3 ± 0.4 5.2 ± 0.7 44.3 ± 1.2
35 MPa n.d. n.d. 3.2 ± 0.4 n.d. 96.8 ± 0.4
Values reported in the table represent the mean ± SD of two independent biological replicates, analyzed in duplicate. 3-OH-C4, 3-
hydroxybutyrate; 3-OH-C12, 3-hydroxydodecanoate; 12-OH-C13, 12-hydroxytridecanoate; 3-OH-C14, 3-hydroxytetradecanoate;
n.d., non-detected.
The production of mcl-PHA monomers by P. denitrificans is, in fact, an interesting

feature. Different PHA biosynthesis pathways are possible in bacteria (Suriyamongkol et al.,
2007). The production of mcl-PHA seems closely linked to fatty acids metabolic routes: β-
oxidation, which is the main metabolic pathway for related substrates (e.g. fatty acids), and de
novo fatty acid biosynthesis with non-related carbon sources (e.g. sugars) (Huijberts et al., 1992).
In the first case, the resulting PHA composition depends on the carbon source, whereas in the
second case, there is no relationship between the carbon sources and the resulting PHA
composition (Możejko-Ciesielska and Kiewisz, 2016). There are several microorganisms
producing mcl-PHA from non-related carbon sources. The most common and well-studied
153
examples are members of the pseudomonas genus (Abe et al., 1994; Huijberts et al., 1992; Kato
et al., 1996; Simon-Colin et al., 2008). However, recent studies report the occurrence of this effect
on different microbial strains. For instance, Shahid et al. (2013) observed production of mcl-PHA
by Bacillus megaterium DSM 509 from unrelated carbon sources, such as glycerol, citrate and
succinate, under nitrogen depletion conditions. The resulting monomer composition of mcl-PHA
related to that of P. putida mt-2, which was unexpected considering that PHA synthases of
Bacillus and Pseudomonas sp. belong to distinct classes regarding size, subunit composition and
substrate specificities. The authors suggested that poly(3-hydroxybutyrate) initially produced was
further degraded, and the energy generated was probably used to drive the metabolic pathways
for mcl-PHA production. Ribeiro et al. (2015) have also reported formation of mcl- and lcl-PHA
(long chain length-PHA) from non-related sources, by Cupriavidus necator IPT 027 and
Burkholderia cepacia IPT 438. The composition of the polymers produced from pure glycerol
predominantly consisted of monomers of 11-hydroxyhexadecanoate and 3-
hydroxytetradecanoate monomers. The authors stated that formation of building blocks and
longer molecular structures of PHA is possibly related to the carbon sources adopted. But the
results in the present study suggest that it can also be affected by other operational conditions,
such as oxygen availability or pressure.
In the 0.1 MPa/C2 samples, the PHA polymer produced was mainly composed by 12-
hydroxytridecanoate, as well as unidentified monomers. Minor components included 3-
hydroxydodecanoate and 3-hydroxytetradecanoate. By applying HP during fermentation, the
PHA composition was found to vary according to the pressure conditions. At 25 MPa, the
monomeric composition was similar to 0.1 MPa/C2, while the same was not observed at 10 MPa.
In the latter, unidentified monomers were the major PHA components, with a minor contribution
of 3-hydroxydodecanoate and 12-hydroxytridecanoate. For some reason, 3-
hydroxytetradecanoate monomers were not even detected in polymer produced at this pressure.
When performing fermentation at 35 MPa, the accumulated polymer was mainly composed by
the unidentified monomers, with a minor proportion of 12-hydroxytridecanoate. With this
information, it is not clear why the polymer composition was modified at 10 and 35 MPa, but not
at 25 MPa, since a progressive and more consistent effect of HP on polymer composition would
be expected. In their study regarding PHA formation by P. putida KT2440, Follonier et al. (2012)
observed no changes in PHA composition when fermentation was carried out at ≈ 0.7 MPa,
compared to the polymer produced at atmospheric pressure. However, it is important to note that
the pressure applied in that work was considerably lower than the pressure levels that we were
dealing in the present study. By using pressure levels in the range of 30 – 100 MPa, Kato et al.
(2007) reported an example of polymer modification under HP conditions, for bacterial cellulose
154
produced by Gluconacetobacter xylinus. Therefore, polymer production at higher pressure levels

seems more prone to promote changes in polymer composition and morphological properties.
Overall, PHA production and composition was highly dependent on the pressure applied
on P. denitrificans growth and fermentation. On the one hand, HP decreased polymer titers, but
increased the PHA content in cell dry mass (%), indicating higher ability to accumulate these
polymers in the cells. On the other hand, some levels of HP affected the PHA monomeric
composition, mainly at 10 and 35 MPa. Although the HP effects on PHA properties were not
evaluated in the present work, it is possible to foresee that the changes in polymer composition
will certainly affect its mechanical properties. For instance, a complex organization of building
blocks with medium and long chains is usually correlated with PHA polymers with higher
molecular weight, and may contribute to their minimal crystallinity and higher industrial
applicability (Laycock et al., 2013; Ribeiro et al., 2016; Simon-Colin et al., 2008). However, in
the particular case of polymers produced under HP, it was not possible to get a specific pattern of
monomer modification, and thus, the effects on mechanical properties are not predictable.
Therefore, future work must focus on further evaluation of the composition of PHA produced
under pressure, as well as determination how it impacts its physical and mechanical properties.
In order to do that, a higher amount of PHA polymers should be obtained, by optimizing
fermentation under HP conditions, and thus increasing the scale of the process to higher volumes,
using a higher volume pressure vessel. In the present study, the experiments were performed on
a HP equipment suitable for pasteurization and food technology purposes, which can also be used
for a wider range of applications, but with some inherent constrains, such as volume limitations
and absence of agitation mechanisms. Therefore, further optimization studies should be
performed in tailor-made HP equipment able to better meet the specifications required for these
microbial growth and fermentation studies, such as agitation. It should be highlighted that such
type of pressure equipment is now becoming available.
8.4. Conclusions
Paracoccus denitrificans growth and fermentation were both affected by HP, with the
effects varying according to the pressure level applied. Despite of the negative impact on cell
growth, interesting metabolic features were observed under HP conditions. For instance, biomass
yields were always lower than at atmospheric pressure, and with a gradual decrease with the
increase of pressure. This indicates that, under HP conditions, P. denitrificans decreased the
utilization of glycerol for purposes of cell growth, suggesting the formation of different metabolic
products. In fact, the PHA content in cell dry mass (%) tended to increase with the increasing
pressure, suggesting higher ability to accumulate these polymers in the cells. High pressure was
155
also found to promote changes in PHA composition, with the polymer produced at 10 and 35 MPa
showing considerable differences relatively to the ones obtained at atmospheric pressure (C1 and
C2). The effects of these modifications on PHA physical and mechanical properties have not yet
been disclosed, and neither the potential improvements arising from that. In any case, the results
obtained from this work demonstrated the possibility of applying HP technology to this type of
fermentation processes, without compromising the production of PHA by P. denitrificans. It also
showed that the PHA produced under HP had different monomeric composition, which may lead
to a polymer with different properties. These results display another example of interesting effects
of pressure on living systems, with possible biotechnological potential. Under the same reasoning,
some pharmaceutical companies are investigating microbial growth under deep-sea like
environments (at HP conditions), aiming to obtain novel compounds with different biological
activities.
To better understand the potential of using HP to change the chemical composition and
mechanical properties of bio-polymers, in particular those produced by P. denitrificans, more
information must be disclosed on this subject. The present study will certainly be helpful in further
optimization studies, as well as in scale-up to higher volumes. In order to do that, tailor-made HP
equipment will be needed, to meet the specifications of these microbial growth and fermentation
processes, which require particular system features, such as the possibility of agitation and air
supply. This type of HP equipment is now becoming more widely available, and allow to easily
apply pressure to any microbial system, during the entire time of growth and/or fermentation, and
with low energetic costs.
156
8.5. References
Abe, H., Doi, Y., Fukushima, T., Eya, H., 1994. Biosynthesis from gluconate of a random
copolyester consisting of 3-hydroxybutyrate and medium-chain-length 3-
hydroxyalkanoates by Pseudomonas sp. 61-3. Int. J. Biol. Macromol. 16, 115–119.
Ashby, R.D., Solaiman, D.K.Y., Foglia, T.A., 2004. Bacterial poly(hydroxyalkanoate) polymer
production from the biodiesel co-product stream. J. Polym. Environ. 12, 105–112.
Braunegg, G., Sonnleitner, B.Y., Lafferty, R.M., 1978. A rapid gas chromatographic method for
the determination of poly-β-hydroxybutyric acid in microbial biomass. Appl. Microbiol.
Davis, R., Kataria, R., Cerrone, F., Woods, T., Kenny, S., O’Donovan, A., Guzik, M., Shaikh, H.,
Duane, G., Gupta, V.K., others, 2013. Conversion of grass biomass into fermentable sugars
and its utilization for medium chain length polyhydroxyalkanoate (mcl-PHA) production by
Pseudomonas strains. Bioresour. Technol. 150, 202–209.
de Almeida, A., Giordano, A.M., Nikel, P.I., Pettinari, M.J., 2010. Effects of aeration on the
synthesis of poly (3-hydroxybutyrate) from glycerol and glucose in recombinant
Escherichia coli. Appl. Environ. Microbiol. 76, 2036–2040.
7997–8002.
Follonier, S., Henes, B., Panke, S., Zinn, M., 2012. Putting cells under pressure: A simple and
efficient way to enhance the productivity of medium-chain-length polyhydroxyalkanoate in
processes with Pseudomonas putida KT2440. Biotechnol. Bioeng. 109, 451–461.
Hori, K., Soga, K., Doi, Y., 1994. Effects of culture conditions on molecular weights of poly (3-
hydroxyalkanoates) produced by Pseudomonas putida from octanoate. Biotechnol. Lett. 16,
709–714.
Huijberts, G.N., Eggink, G., De Waard, P., Huisman, G.W., Witholt, B., 1992. Pseudomonas
putida KT2442 cultivated on glucose accumulates poly(3-hydroxyalkanoates) consisting of
saturated and unsaturated monomers. Appl. Environ. Microbiol. 58, 536–544.
157
Kato, M., Bao, H.J., Kang, C.-K., Fukui, T., Doi, Y., 1996. Production of a novel copolyester of
3-hydroxybutyric acid and medium-chain-length 3-hydroxyalkanoic acids by Pseudomonas
sp. 61-3 from sugars. Appl. Microbiol. Biotechnol. 45, 363–370.
698.
Kenny, S.T., Runic, J.N., Kaminsky, W., Woods, T., Babu, R.P., Keely, C.M., Blau, W.,
O’Connor, K.E., 2008. Up-cycling of PET (polyethylene terephthalate) to the biodegradable
plastic PHA (polyhydroxyalkanoate). Environ. Sci. Technol. 42, 7696–7701.
Kumar, P., Jun, H.-B., Kim, B.S., 2018. Co-production of polyhydroxyalkanoates and carotenoids
through bioconversion of glycerol by Paracoccus sp. strain LL1. Int. J. Biol. Macromol.
107, 2552–2558.
Laycock, B., Halley, P., Pratt, S., Werker, A., Lant, P., 2013. The chemomechanical properties of
microbial polyhydroxyalkanoates. Prog. Polym. Sci. 38, 536–583.
Eng. Life Sci. 7, 475–479.
Możejko-Ciesielska, J., Kiewisz, R., 2016. Bacterial polyhydroxyalkanoates: Still fabulous?
Microbiol. Res. 192, 271–282.
426–433.
158

Ribeiro, P.L.L., da Silva, A.C.M.S., Menezes Filho, J.A., Druzian, J.I., 2015. Impact of different
by-products from the biodiesel industry and bacterial strains on the production, composition,
and properties of novel polyhydroxyalkanoates containing achiral building blocks. Ind.
Crops Prod. 69, 212–223.
Ribeiro, P.L.L., Souza Silva, G., Druzian, J.I., 2016. Evaluation of the effects of crude glycerol
on the production and properties of novel polyhydroxyalkanoate copolymers containing
high 11-hydroxyoctadecanoate by Cupriavidus necator IPT 029 and Bacillus megaterium
IPT 429. Polym. Adv. Technol. 27, 542–549.
Shahid, S., Mosrati, R., Ledauphin, J., Amiel, C., Fontaine, P., Gaillard, J.-L., Corroler, D., 2013.
Impact of carbon source and variable nitrogen conditions on bacterial biosynthesis of
polyhydroxyalkanoates: evidence of an atypical metabolism in Bacillus megaterium DSM
509. J. Biosci. Bioeng. 116, 302–308.
Simon-Colin, C., Raguénès, G., Costa, B., Guezennec, J., 2008. Biosynthesis of medium chain
length poly-3-hydroxyalkanoates by Pseudomonas guezennei from various carbon sources.
React. Funct. Polym. 68, 1534–1541.
Suriyamongkol, P., Weselake, R., Narine, S., Moloney, M., Shah, S., 2007. Biotechnological
approaches for the production of polyhydroxyalkanoates in microorganisms and plants - A
review. Biotechnol. Adv. 25, 148–175.
Tan, G.-Y.A., Chen, C.-L., Li, L., Ge, L., Wang, L., Razaad, I.M.N., Li, Y., Zhao, L., Mo, Y.,
Wang, J.-Y., 2014. Start a research on biopolymer polyhydroxyalkanoate (PHA): A review.
Polymers (Basel). 6, 706–754.
Üçisik-Akkaya, E., Ercan, O., Yesiladali, S.K., Öztürk, T., Ubay-Çokgör, E., Orhon, D.,
Tamerler, C., Çakar, Z.P., 2009. Enhanced polyhydroxyalkanoate production by
Paracoccus pantotrophus from glucose and mixed substrate. Fresenius Environ. Bull. 18,
2013–2022.
3574–3579.
159
160
CHAPTER IX
Conclusions and outlook

CHAPTER IX. Conclusions and outlook
Application of sub-lethal HP has a tremendous potential to modulate (and potentially

improve) glycerol-based fermentation processes. In the present work, the effects of pressure were
evaluated on two microbial strains, involved in two different fermentation processes. The most
relevant distinct features between them relied on their oxygen requirements and on their HP
tolerance. Regarding the oxygen requirements, the results showed that it was possible to perform
fermentation under pressure at both anaerobic and aerobic conditions, such as observed for L.
reuteri and P. denitrificans, respectively. However, many high pressure equipments currently
available (including the one used in the present work) are not adapted to allow continuous air
supply and agitation. Therefore, this approach is generally more suitable for anaerobic processes,
which do not have high oxygen requirements, and, because of that, most studies on this field are
carried out with anaerobic or facultative anaerobic microbial strains. The other distinctive factor
between the two microbial strains was their tolerance to HP. Lactobacillus reuteri and P.
denitrificans were both able to grow at low pressure levels (e.g. 10 or 25 MPa), but showed
different tolerance to higher pressure levels. While P. denitrificans was still able to grow at 35
MPa, L. reuteri was already inhibited at this pressure. This may indicate that both strains have
different response mechanisms to HP and/or that their cellular and molecular structures have
different resistance to pressure. Despite all the differences between the fermentative strains and
respective processes, sub-lethal HP was successfully applied throughout fermentation. By
optimization of pressure and general fermentative conditions, the microbial strains were able to
grow and retain the fermentative activity, while promoting interesting metabolic changes in both
cases.
For the studies with L. reuteri under HP, some of the most important findings are
indicated as follows:
- High pressure (10 – 35 MPa) affected L. reuteri growth and fermentation, with the effects
varying according to the pressure level and the initial acetate content. In general,
fermentation was less inhibited by HP when acetate was present, indicating that acetate
enhances the resistance of L. reuteri to pressure.
- Production of 1,3-PDO was stimulated at 10 MPa, leading to higher titers, yields and
productivities, compared to 0.1 MPa. In fact, fermentation at 10 MPa promoted a
metabolic shift, with modification of product selectivity towards production of 1,3-PDO,
and general reduction in the formation of by-products.
- Application of consecutive fermentation cycles increased 1,3-PDO production, especially
at 10 MPa, in samples without added acetate. In that case, considerable improvements of
1,3-PDO titers were achieved relatively to the “conventional approach”, i.e. without the
fermentation cycles and at 0.1 MPa.
163
- At 25 MPa, 1,3-PDO production was also improved between the first and the fourth
cycles, but the increment was less pronounced and highly variable, showing an
“improvement and decline” trend. This suggests that L. reuteri was not able to adapt (at
least not steadily) to a pressure of 25 MPa, within the number of cycles studied.
- The comparative metabolomic study between L. reuteri fermentation samples at 0.1 and
10 MPa showed a modification in metabolic profiles throughout the cycles. This effect
was considerably more accentuated at 10 MPa. One of the metabolites characteristic of
fermentation cycles at 10 MPa was 2,3-BDO, which showed higher relative abundance
at this pressure, relatively to the respective cycle at 0.1 MPa.
The most relevant observations concerning the effects of air availability and HP on P.
denitrificans are also summarized below:
- Paracoccus denitrificans growth and metabolism were affected by air availability. At
higher air availability, considerable cell growth was observed, but no production of
ethanol, acetic or succinic acids. Without air availability, P. denitrificans showed active
metabolic activity (with production of ethanol, acetic and succinic acids), but no cell
growth.
- Paracoccus denitrificans was able to grow at 10 and 25 MPa, with biomass
concentrations similar to 0.1 MPa. A more extensive inhibitory effect occurred at 35 MPa,
while no cell growth was observed at 50 MPa. HP also affected substrate consumption,
as well as the formation of ethanol, acetic and succinic acids, with the fermentative profile
varying according to the pressure level.
- High pressure decreased PHA titers, but increase the PHA content in cell dry mass,
suggesting higher ability to accumulate these polymers in the cells. In addition, HP
promoted changes in PHA monomeric composition, with the polymer produced at 10 and
35 MPa showing considerable differences relative to the ones obtained at atmospheric
pressure.
Overall, the present work indicates that HP can be applied to different microorganisms,
with different process specificities. The combination of these studies provide a general overview
of the main potentialities and limitations of HP application to glycerol-based processes. The final
titers of 1,3-PDO and PHA obtained with this strategy were low compared to some other studies
in literature, and far from those required for industrial application. However, these results unveil
relevant information regarding the adaptation of mesophilic microorganisms to sub-lethal HP,
and give a perspective on how these mechanisms can be used to stimulate or inhibit specific
metabolic pathways, similarly to genetic engineering approaches, but with a more general and
164
less targeted approach. Moreover, the insights provided by this work pave the way for further
optimization studies, with HP equipment suitable for these microbial growth and fermentation
approaches. This will promote the use of these type of strategies on a wide range of microbial
processes, with potential application in the food, pharmaceutical and energy industries.
165
166
APPENDIX A
Additional information supporting

CHAPTER VI
Appendix A
Figure A.1. Examples of 1H NMR spectra of fermentation samples after one and four cycles at atmospheric
pressure (C1 and C4, respectively), and at 10 MPa (P1 and P4, respectively): (a) full spectra; and expansions
for (b) aromatic region (6.0-9.0 ppm) and (c) aliphatic region (0.6-3.0 ppm).
169
Appendix A
Figure A.2. Principal component analysis (PCA) loading plots PC1 (a) and PC2 (b) of L. reuteri
fermentation metabolites, obtained by 1D 1H NMR.
170
Appendix A
Figure A.3. Metabolite plots showing the abundance of 1,3-PDO (for M44 and M60, both signals
corresponding to this metabolite) in samples at 0.1 MPa (C1 and C4, for the first and fourth fermentation
cycles, respectively) and at 10 MPa (P1 and P4 samples, for the first and fourth fermentation cycles,
respectively).
171
Appendix A
172
APPENDIX B
Additional information supporting

CHAPTER VII
Adapted from:
Mota, M.J., Lopes, R.P., Pinto, C.A, Sousa, S., Gomes, A.M., Delgadillo, I., Saraiva, J.A., The
use of different fermentative approaches on Paracoccus denitrificans: Effect of high pressure and
air availability on growth and metabolism. Submitted in Biotechnology Progress.
Appendix B
Table B.1. Glucose and maltose concentrations, at 0 h and 72 h, for fermentation under different
air availability conditions.
Maltose concentration (g L-1) Glucose concentration (g L-1)

Samples
0h 72 h 0h 72 h
Without air n.d. 1.126 ± 0.005
24h with air + 48h

0.256 ± 0.024 n.d. 1.066 ± 0.012 1.037 ± 0.063
without air
With air n.d. 1.155 ± 0.085
Values reported in the table represent the mean ± SD of two independent biological replicates, analyzed in duplicated. N.d. indicates
non-detected production of the compound.
175

Microbial Fermentation of Glycerol For Production of Value-Added Compounds Under High Pressure

Caricato da

Informazioni sul documento

Titolo originale

Copyright

Formati disponibili

Condividi questo documento

Condividi o incorpora il documento

Opzioni di condivisione

Hai trovato utile questo documento?

Questo contenuto è inappropriato?

Copyright:

Formati disponibili

Microbial Fermentation of Glycerol For Production of Value-Added Compounds Under High Pressure

Caricato da

Copyright:

Formati disponibili

Universidade de Aveiro Departamento de Química

MARIA JOÃO PINHO FERMENTAÇÃO MICROBIANA DE GLICEROL PARA

MICROBIAL FERMENTATION OF GLYCEROL FOR

MARIA JOÃO PINHO FERMENTAÇÃO MICROBIANA DE GLICEROL PARA

MICROBIAL FERMENTATION OF GLYCEROL FOR

Apoio financeiro do POCTI no âmbito Apoio financeiro da FCT e do FSE no

Ao meu orientador, Professor Jorge Saraiva, pela disponibilidade, apoio e por

À minha co-orientadora, Professora Ivonne Delgadillo, pela orientação e

À Professora Ana Gomes e ao Sérgio Sousa, pelo apoio no HPLC; ao

A todos os colegas do Innovate Group, pela colaboração no desenvolvimento

Em particular, à Ana Alves e à Betânia, pela amizade sincera, independente do

E, por fim, ao Zinho, pelo amor, amizade e compreensão, e pelo apoio

resumo A conversão microbiana de glicerol em compostos de valor-acrescentado

abstract Microbial conversion of glycerol into value-added products constitutes a

List of Figures ............................................................................................................................... i

Chapter I. Fermentation at non-conventional conditions and its application to food- and

Chapter II. Fermentation of glycerol for production of valuable bio-chemicals: A

Chapter III. Scope and outline of the work ............................................................................ 45

Chapter V. Utilization of glycerol during consecutive cycles of Lactobacillus reuteri

Chapter VI. Comparative metabolomic profiling of Lactobacillus reuteri under high

Chapter VII. The use of different fermentative approaches on Paracoccus denitrificans:

Chapter VIII. Effect of high pressure on Paracoccus denitrificans growth and

Chapter IX. Conclusions and outlook................................................................................... 161

Appendix A. Additional information supporting Chapter VI ........................................... 167

Appendix B. Additional information supporting Chapter VII .......................................... 173

Figure 6.5. Schematic representation of general metabolic pathways for 2,3-butanediol

Figure 7.4. Glucose concentrations throughout fermentation at different HP conditions: 10 MPa,

Table 8.2. Polyhydroxyalkanoate contents, yields and productivities, after 72 h of fermentation,

Table 8.3. Polyhydroxyalkanoate monomeric composition (mol%), after 72 h of fermentation, at

This thesis is based on the following scientific articles:

Fermentation at non-conventional conditions in food- and bio-sciences by application of

Utilization of glycerol during consecutive cycles of Lactobacillus reuteri fermentation under

The use of different fermentative approaches on Paracoccus denitrificans: Effect of high

Effect of high pressure on Paracoccus denitrificans growth and polyhydroxyalkanoates

Effect of ultrasound on lactic acid production by Lactobacillus strains in date (Phoenix

Application of High Pressure with Homogenization, Temperature, Carbon Dioxide, and

Hyperbaric storage at and above room temperature of a highly perishable food

Fermentation at non-conventional conditions

1.1. General overview

1.2. Fermentation under high pressure conditions

Microorganism Main effects References

Thermophilic bacterium Clostridium thermocellum efficiently converts cellobiose to

1.2.1. Technological equipment suitable for fermentation processes

1.3. Comparison between high pressure and other non-conventional technologies

laboratory-scale, the scale-up of these processes will be an important requirement to their

pulsed electric fields: An engineering perspective. Food Eng. Rev. 3, 94–107.

Fermentation of glycerol for production of

2.1. The biodiesel industry and the glycerol surplus

2.2. Glycerol bio-conversion into valuable products

Bioproduct Main producing-microorganisms References

Butanol 0.50 0.58

Although glycerol conversion is highly variable according to the microorganism, it is

The reductive pathway proceeds by the dehydration of glycerol to 3-

2.3. Glycerol bio-conversion into 1,3-propanediol

2.3.1. Production of 1,3-propanediol by Lactobacillus reuteri

Titer (g L-1) Yield (g g-1) Productivity (g L-1 h-1) References

Although there is no apparent straightforward strategy to overcome the former limitation,

2.4. Glycerol bio-conversion into polyhydroxyalkanoates

2.4.1. Polyhydroxyalkanoate production by Paracoccus denitrificans

Titer Yield Productivity Content

Kalaiyezhini and Ramachandran (2015) have also studied poly(3-hydroxybutyrate)

2.5. Current challenges and possible improvements

Scope and outline of the work